A tower tilt monitoring method and device based on camera internet network

Abstract

A tower inclination monitoring method and device based on a camera Internet network, belonging to the technical field of wind turbine tower monitoring, the method measures and records the actual height of the target tower; according to the actual height and the surrounding environment, a camera is placed at the center of the adjacent tower top; the data of the target tower at this time is used as initialization data, and the target tower is photographed in real time; the real-time image and the projection line laser image are processed respectively; the inclination angle and settlement data are calculated according to the edge line of the target tower and the laser center straight line; the inclination angle of the target tower is compared and calculated with the inclination angle of the reference wind turbine, and if the difference between the two exceeds the preset safety threshold, an alarm is issued; the settlement data of the target tower is compared with the initialization data, and if a height settlement occurs, an alarm is issued. The present invention can realize all-round and multi-angle monitoring of wind turbine towers, automatically identify and judge the degree of inclination of wind turbine towers, and reduce manual intervention.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbine tower monitoring, and in particular to a tower tilt monitoring method and device based on a camera internet network. Background Art

[0002] The wind turbine tower is the core supporting structure of the wind turbine and bears the weight of the entire wind turbine. The tower is usually connected by multiple sections of threads, and each section of the tower is connected by high-strength bolts. These bolts may become loose, fatigue damaged, corroded, etc. when subjected to complex stress environments such as wind loads and heavy loads for a long time.

[0003] The annual inspection of wind turbine tower bolts is one of the important maintenance measures in the field of wind power generation. The loosening and damage of wind turbine tower bolts may lead to the decrease of wind turbine tower structure stability, increase maintenance costs and downtime. In serious cases, it may cause safety accidents and cause damage to personnel and equipment.

[0004] At present, the annual inspection of wind turbine tower bolts is all done manually with mechanical assistance, and there are several major problems:

[0005] First, the accuracy is insufficient: the existing GPS technology and sensor measurements cannot fully meet the requirements of millimeter-level tilt measurement of wind turbine towers, resulting in errors in the measurement results.

[0006] Second, the cost is high: the use of high-precision measurement equipment and technology is often accompanied by higher costs, which increases the investment burden of wind power projects.

[0007] Third, insufficient consideration is given to nonlinear deformation: existing calculation methods are often based on simplified assumptions and do not fully consider the nonlinear deformation characteristics of wind turbine towers, resulting in inaccurate displacement calculation results.

[0008] Fourth, the benchmark point problem: the benchmark points in wind farms are susceptible to environmental erosion and wear, which affects the accuracy of measurements. Early wind farms may not have benchmark points, resulting in the inability to fully reflect the accumulated settlement of the wind turbine foundation.

[0009] Therefore, how to invent a method for monitoring the inclination of wind turbine towers that can achieve all-round and multi-angle monitoring of wind turbine towers, automatically identify and judge the degree of inclination of wind turbine towers, thereby reducing the need for manual intervention has become an urgent problem to be solved. Summary of the invention

[0010] To this end, the present invention provides a tower tilt monitoring method and device based on a camera Internet network, which can realize all-round and multi-angle monitoring of wind turbine towers, automatically identify and judge the tilt degree of wind turbine towers, thereby reducing the need for manual intervention.

[0011] In order to achieve the above-mentioned object, the present invention provides the following technical solution: a tower inclination monitoring method based on a camera interconnection network, comprising:

[0012] Measuring the height of a target tower and recording the measured actual height of the target tower;

[0013] According to the actual height of the target tower and the surrounding environment, a camera is placed at the center of the top of the adjacent tower to ensure that the target tower can be imaged by a set number of adjacent towers;

[0014] The real-time data of the target tower obtained by a camera placed at the center of the top of the adjacent tower is used as initialization data; after the initialization data is set, the target tower is photographed in real time by the camera to obtain and save the real-time image of the target tower;

[0015] The real-time image of the target tower taken by the camera is transmitted to the image processing module of the camera Internet network; the real-time image of the target tower and the projection line laser image are processed by the image processing module respectively to extract the edge line of the target tower and the laser center line;

[0016] A three-dimensional coordinate system is established according to the center line of the target tower, the wind turbine and the wind turbine impeller; in the three-dimensional coordinate system, the inclination angle and settlement data of the target tower are calculated by extracting the edge line of the target tower and the laser center straight line;

[0017] The inclination angle of the target tower is compared with the inclination angle of the reference wind turbine. If the difference between the two exceeds a preset safety threshold, an alarm is issued; the settlement data of the target tower is compared with the initialization data of the target tower. If a high settlement occurs, an alarm is issued.

[0018] As a preferred solution of a tower inclination monitoring method based on a camera interconnection network, in order to ensure that the target tower can be imaged by a set number of adjacent towers, a camera is installed on the tower using a camera installation algorithm; the camera is calculated according to the algorithm as follows:

[0019]

[0020] Where C is the total number of cameras required; m and n represent the number of wind turbines in the horizontal and vertical directions, respectively; k is the number of additional cameras due to terrain, obstacles or specified environmental factors; f(m,n) is a correction function used to take into account the irregularity of the layout and other influencing factors.

[0021] As a preferred solution of a tower tilt monitoring method based on a camera interconnection network, the camera in the camera interconnection network can maintain stable operation through a Nash equilibrium camera interconnection algorithm; the expression of the Nash equilibrium camera interconnection algorithm is:

[0022]

[0023] in, Represents C i The optimal solution for tracking the target is a specific camera C i ∈C, and Represents the set of all cameras.

[0024] As a preferred solution of a tower inclination monitoring method based on a camera Internet network, in the process of comparing and calculating the inclination angle of the target tower with the inclination angle of the reference wind turbine, the calculation formula of the inclination angle of the reference wind turbine is:

[0025]

[0026] In the formula, θ ref is the inclination angle of the reference fan; h top and h bottom are the heights of the top and bottom of the wind turbine tower respectively, and d is the distance from the bottom to the top of the tower.

[0027] As a preferred solution of a tower inclination monitoring method based on a camera interconnection network, in the process of respectively processing the real-time image and the projection line laser image of the target tower by the image processing module, extracting the edge line and the laser center line of the target tower, the real-time image of the target tower is grayed by the weighted average method and the maximum value method;

[0028] The calculation formula of the weighted average method is:

[0029] Gray(x, y) = C R R(x,y)+C G G(x,y)+C B B(x,y)

[0030] In the formula, the coefficient C R , C G , C B Represent the weights of the three channels of the image, all greater than zero, and C R +C G +C B =1; (x, y) is the coordinate of the pixel point in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel points in the R, G and B channels respectively;

[0031] The calculation formula of the maximum value method is:

[0032] Gray(x,y)=max(R(x,y)+max(G(x,y)+B(x,y)))

[0033] Where (x, y) is the coordinate of the pixel in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel in the R, G and B channels respectively.

[0034] As a preferred solution of a tower inclination monitoring method based on a camera Internet network, in the process of respectively processing the real-time image and the projection line laser image of the target tower by the image processing module and extracting the edge line and the laser center line of the target tower, the real-time image of the target tower that has completed the grayscale processing is subjected to noise reduction processing by Gaussian filtering, and the calculation formula of the Gaussian function with two-dimensional zero mean and standard deviation σ is:

[0035]

[0036] Where x and y are the coordinates of the image pixel points; σ is the standard deviation. The larger the σ value is, the smaller the difference between the different coefficient values ​​in the generated Gaussian template is, and the more obvious the smoothing effect on the image is.

[0037] The present invention also provides a tower inclination monitoring device based on a camera Internet network, based on the above tower inclination monitoring method based on a camera Internet network, comprising:

[0038] The actual tower height measurement module is used to measure the height of the target tower and record the measured actual height of the target tower;

[0039] A camera layout placement module is used to place cameras at the center of the top of adjacent towers according to the actual height of the target tower and the surrounding environment, so as to ensure that the target tower can be imaged by a set number of adjacent towers;

[0040] A real-time image acquisition module is used to use the real-time data of the target tower obtained by a camera placed at the center of the top of the adjacent tower as initialization data; after the initialization data is set, the target tower is photographed in real time by the camera to obtain and save the real-time image of the target tower;

[0041] An image data processing module is used to transmit the real-time image of the target tower taken by the camera to the image processing module of the camera Internet network; the real-time image of the target tower and the projection line laser image are processed by the image processing module respectively to extract the edge line of the target tower and the laser center line;

[0042] A tower real-time tilt angle and settlement data acquisition module is used to establish a three-dimensional coordinate system according to the center line of the target tower, the fan and the wind turbine impeller; in the three-dimensional coordinate system, the tilt angle and settlement data of the target tower are calculated by extracting the edge line of the target tower and the laser center straight line;

[0043] The data comparison alarm module is used to compare and calculate the inclination angle of the target tower with the inclination angle of the reference wind turbine. If the difference between the two exceeds a preset safety threshold, an alarm reminder is issued; the settlement data of the target tower is compared with the initialization data of the target tower. If a high settlement occurs, an alarm reminder is issued.

[0044] As a preferred solution of a tower inclination monitoring device based on a camera interconnection network, in the camera layout placement module, in order to ensure that the target tower can be imaged by a set number of adjacent towers, a camera is installed on the tower through a camera installation algorithm; the camera is calculated according to the algorithm as follows:

[0045]

[0046] Where C is the total number of cameras required; m and n represent the number of wind turbines in the horizontal and vertical directions, respectively; k is the number of additional cameras due to terrain, obstacles or specified environmental factors; f(m,n) is a correction function used to take into account the irregularity of the layout and other influencing factors;

[0047] In the camera layout placement module, the Nash equilibrium camera interconnection algorithm is also used to enable the cameras in the camera interconnection network to maintain stable operation; the expression of the Nash equilibrium camera interconnection algorithm is:

[0048]

[0049] in, Represents C i The optimal solution for tracking the target is a specific camera C i ∈C, and Represents the set of all cameras.

[0050] As a preferred solution of a tower inclination monitoring device based on a camera Internet network, in the data comparison alarm module, in the process of comparing and calculating the inclination angle of the target tower with the inclination angle of the reference wind turbine, the calculation formula of the inclination angle of the reference wind turbine is:

[0051]

[0052] In the formula, θ ref is the inclination angle of the reference fan; htop and h bottom are the heights of the top and bottom of the wind turbine tower respectively, and d is the distance from the bottom to the top of the tower.

[0053] As a preferred solution of a tower inclination monitoring device based on a camera interconnection network, in the image data processing module, in the process of processing the real-time image and the projection line laser image of the target tower by the image processing module respectively, extracting the edge line of the target tower and the laser center line, the real-time image of the target tower is grayed by the weighted average method and the maximum value method;

[0054] The calculation formula of the weighted average method is:

[0055] Gray(x, y) = C R R(x,y)+C G G(x,y)+C B B(x,y)

[0056] In the formula, the coefficient C R , C G , C B Represent the weights of the three channels of the image, all greater than zero, and C R +C G +C B =1; (x, y) is the coordinate of the pixel point in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel points in the R, G and B channels respectively;

[0057] The calculation formula of the maximum value method is:

[0058] Gray(x,y)=max(R(x,y)+max(G(x,y)+B(x,y)))

[0059] Where (x, y) is the coordinate of the pixel point in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel points in the R, G and B channels respectively;

[0060] After the real-time image of the target tower is grayed, the real-time image of the target tower that has been grayed is subjected to noise reduction processing by Gaussian filtering. The calculation formula of the Gaussian function with two-dimensional zero mean and standard deviation σ is:

[0061]

[0062] Where x and y are the coordinates of the image pixel points; σ is the standard deviation. The larger the σ value is, the smaller the difference between the different coefficient values ​​in the generated Gaussian template is, and the more obvious the smoothing effect on the image is.

[0063] The present invention has the following advantages: measuring the height of a target tower and recording the actual height of the target tower obtained by measurement; placing a camera at the center of the top of an adjacent tower according to the actual height of the target tower and the surrounding environment to ensure that the target tower can be imaged by a set number of adjacent towers; using the real-time data of the target tower acquired by the camera placed at the center of the top of the adjacent tower as initialization data; after the initialization data is set, photographing the target tower in real time through the camera to acquire and save the real-time image of the target tower; transmitting the real-time image of the target tower photographed by the camera to an image processing module of a camera Internet network; and The module processes the real-time image and the projection line laser image of the target tower respectively, extracts the edge line and the laser center line of the target tower; establishes a three-dimensional coordinate system according to the center line of the target tower, the fan and the wind turbine impeller; in the three-dimensional coordinate system, calculates the inclination angle and settlement data of the target tower according to the extracted edge line and the laser center line of the target tower; compares and calculates the inclination angle of the target tower with the inclination angle of the reference fan, and if the difference between the two exceeds the preset safety threshold, an alarm is issued; compares the settlement data of the target tower with the initialization data of the target tower, and if a high settlement occurs, an alarm is issued. The present invention can realize all-round and multi-angle monitoring of wind turbine towers by using a multi-camera network. Through the collaborative work between multiple cameras, the inclination of the wind turbine tower can be more comprehensively understood, and the reliability and accuracy of the detection can be improved. Through advanced image acquisition equipment, the original image processing can capture high-definition images of the wind turbine tower, thereby retaining more detailed information. The camera network is used to detect the inclination of the wind turbine tower, realizing non-contact monitoring. This method does not require the installation of sensors or other equipment on the wind turbine tower, which reduces interference with wind turbine equipment and also reduces installation and maintenance costs. Combining computer vision and artificial intelligence technology, the images captured by the camera can be intelligently analyzed to automatically identify and determine the degree of inclination of the wind turbine tower. This intelligent analysis improves the accuracy and efficiency of detection and reduces the need for manual intervention. Through the camera network, the inclination of the wind turbine tower can be displayed to managers or relevant personnel in real time in the form of images or videos. This visual display helps to intuitively understand the operating status of the wind turbine tower, making it easier to find problems in a timely manner and take corresponding measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0065] The structures, proportions, sizes, etc. illustrated in this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with the technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantial technical significance. Any structural modification, change in proportion or adjustment of size shall still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0066] Figure 1 A schematic flow chart of a tower inclination monitoring method based on a camera Internet network provided in Example 1 of the present invention;

[0067] Figure 2 A schematic diagram of the structure of an existing machine vision application system in a tower inclination monitoring method based on a camera interconnection network provided in Example 1 of the present invention;

[0068] Figure 3 A schematic diagram of camera placement in a tower inclination monitoring method based on a camera interconnection network provided in Example 1 of the present invention;

[0069] Figure 4 Schematic diagram of a camera grid in a tower tilt monitoring method based on a camera interconnection network provided in Example 1 of the present invention, wherein (a) is a 4*4 grid; (b) is a 5*5 grid;

[0070] Figure 5 This is a schematic diagram of the architecture of a tower inclination monitoring device based on a camera Internet network provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0071] The following is a description of the implementation of the present invention by specific embodiments. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0072] Example 1

[0073] See also Figure 1Embodiment 1 of the present invention provides a tower inclination monitoring method based on a camera interconnection network, comprising the following steps:

[0074] S1. Measuring the height of a target tower and recording the actual height of the target tower obtained by measurement;

[0075] S2. According to the actual height of the target tower and the surrounding environment, a camera is placed at the center of the top of the adjacent tower to ensure that the target tower can be imaged by a set number of adjacent towers;

[0076] S3, using the real-time data of the target tower obtained by a camera placed at the center of the top of the adjacent tower as initialization data; after the initialization data is set, the target tower is photographed in real time by the camera, and the real-time image of the target tower is obtained and saved;

[0077] S4, transmitting the real-time image of the target tower taken by the camera to the image processing module of the camera Internet network; processing the real-time image of the target tower and the projection line laser image respectively by the image processing module, and extracting the edge line of the target tower and the laser center line;

[0078] S5, establishing a three-dimensional coordinate system according to the center line of the target tower, the wind turbine and the wind turbine impeller; in the three-dimensional coordinate system, calculating and obtaining the inclination angle and settlement data of the target tower according to the edge line of the target tower and the laser center straight line;

[0079] S6. Compare and calculate the inclination angle of the target tower with the inclination angle of the reference wind turbine. If the difference between the two exceeds a preset safety threshold, an alarm is issued; compare the settlement data of the target tower with the initialization data of the target tower. If a high settlement occurs, an alarm is issued.

[0080] In this embodiment, in step S1, the height of the target tower is measured, and the actual height of the target tower obtained by measurement is recorded;

[0081] Specifically, the actual height of the target tower is measured and recorded using professional measuring equipment.

[0082] In this embodiment, in step S2, according to the actual height of the target tower and the surrounding environment, a camera is placed at the center of the top of the adjacent tower to ensure that the target tower can be imaged by a set number of adjacent towers;

[0083] Among them, see Figure 2, a basic machine vision application system consists of a light source, a lens, an industrial camera, and an image processing module; in industrial applications, it also includes a decision-making module and a control execution module. The most critical parts of a machine vision system are the lens and the industrial camera. The industrial camera converts the optical signal into an electrical signal for analysis and processing by the image processing module. The industrial camera requires higher anti-interference ability and data transmission capability. Choosing a suitable industrial camera is an important part of the design of a machine vision system. The industrial camera directly determines the quality of the collected image and determines whether the entire system can run smoothly.

[0084] The image processing module is the key to the camera interconnection system. Image processing is used to complete the recognition, measurement, and detection of the object being measured. Common image processing methods include image enhancement, image sharpening, image segmentation, and feature information extraction. Edge extraction, image fusion, Hough transform, and other algorithms are widely used in image processing in camera interconnection systems.

[0085] Specifically, Figure 3 As shown, according to the actual tower height and surrounding environment, another tower O 2 Place the device. 2 The point is located at the center of the tower top, ensuring that each tower is imaged by at least two adjacent towers.

[0086] When installing cameras, the layout of the cameras needs to be fully considered. Under the premise of ensuring comprehensive monitoring of wind turbines, the number of cameras should be minimized to reduce costs. Therefore, the present invention proposes a professional algorithm based on graph theory and optimization theory. The construction of the algorithm is based on the following premises:

[0087] Wind turbines are arranged in a regular m×n matrix, where m and n represent the number of wind turbines in the horizontal and vertical directions, respectively. In the initialization phase, the algorithm first defines the size of the wind turbine layout matrix and performs special processing on edge nodes to ensure that these nodes can be effectively covered.

[0088] In this embodiment, the camera installation core algorithm adopts an internal node coverage strategy. This strategy achieves full coverage of nodes by deploying cameras at suitable locations. In addition, the algorithm also considers the optimization of camera deployment, reducing the total number of cameras by reducing the overlapping coverage area of ​​cameras, thereby maximizing cost-effectiveness.

[0089] The camera is calculated according to the algorithm formula:

[0090]

[0091] Where C is the total number of cameras required; m and n represent the number of wind turbines in the horizontal and vertical directions, respectively; k is the number of additional cameras due to terrain, obstacles or specified environmental factors; f(m,n) is a correction function used to take into account the irregularity of the layout and other influencing factors.

[0092] The correction function f(m,n) can be defined according to the actual layout and environmental conditions of the wind turbine. The following are some possible considerations:

[0093] Layout irregularity: If the layout of wind turbines is not a strict rectangular grid, a parameter α related to the layout irregularity can be introduced, whose value depends on the degree of deviation of the layout.

[0094] Environmental factors: Considering factors such as terrain undulation, vegetation coverage, and building obstruction, the environmental impact factor β is introduced.

[0095] Wind turbine size: Wind turbines of different sizes may require different numbers of cameras for effective coverage, introducing a size factor γ.

[0096] Camera performance: Considering the camera’s viewing angle, resolution, and observation range, a performance factor δ is introduced.

[0097] Monitoring accuracy requirements: According to the accuracy requirements of the monitoring system, the accuracy factor∈ is introduced.

[0098] Based on the above factors, the correction function f(m,n) can be defined as:

[0099] f(m,n)=α*g(m,n)+β*h(m,n)+γ*s+δ*p+∈*q

[0100] Where g(m,n) is a function related to layout irregularity; h(m,n) is a function related to environmental factors; s is the average size of the wind turbine; p represents the average performance index of the camera; and q represents the required level of monitoring accuracy.

[0101] like Figure 4 As shown in the figure, the camera installation algorithm uses the point coverage problem theory in graph theory and combines heuristic or meta-heuristic algorithms to optimize the camera layout to achieve the optimal allocation of costs and resources while meeting the monitoring accuracy requirements. In the verification and testing phase, the effectiveness of the algorithm was verified through simulation and field testing to ensure that each wind turbine node was covered by at least two cameras.

[0102] Different installation terrains have a significant impact on camera layout, especially when calculating the number of cameras, terrain factors will directly affect the parameters and correction functions in the formula. The specific impact of different installation terrains on the specific values ​​of the parameters in the formula in this embodiment is as follows:

[0103] 1. Flat terrain

[0104] In flat terrain, the layout of wind turbines is usually more regular, and the coverage and angle of the camera can be distributed more evenly. Therefore, in a flat terrain environment, the parameters in the correction function f(m,n) are as follows:

[0105] α (Layout Irregularity Coefficient): The value is smaller because the layout irregularity is lower;

[0106] β (environmental factor coefficient): The value is also small, and flat terrain usually means less environmental interference;

[0107] γ (wind turbine size factor): needs to be adjusted according to the specific size of the turbine, but the overall impact is small;

[0108] δ (Camera Performance Coefficient): The value is moderate, because flat terrain does not require particularly high camera performance;

[0109] Number of cameras: Due to the small values ​​of k (the number of additional cameras) and f(m,n) (the correction function), the number of cameras required is close to the result calculated by the basic formula C = 2(m+n)-4.

[0110] 2. Hilly terrain

[0111] Hilly terrain has large fluctuations, which may affect the coverage and angle of the camera. Therefore, in a hilly terrain environment, the parameters in the correction function f(m,n) are as follows:

[0112] α: Larger values ​​because the terrain irregularity is higher and more cameras are needed to cover the blind spots;

[0113] β: Large value, because the terrain may cause the camera's viewing angle and coverage to change;

[0114] γ: The layout of wind turbines may need to be adjusted according to the terrain, thus affecting the camera coverage requirements;

[0115] δ: Larger values ​​because higher performance cameras are needed to adapt to complex terrain;

[0116] Number of cameras: As k and f(m,n) increase, the number of cameras required will be higher than that on flat terrain to ensure that there are no blind spots.

[0117] 3. Mountainous terrain

[0118] Mountainous terrain is the most complex, and the layout of the camera needs to take into account the undulations and occlusions of the terrain. Therefore, in a mountainous terrain environment, the parameters in the correction function f(m,n) are as follows:

[0119] α: The value is very large because the irregularity of mountain terrain is extremely high, and the layout of the cameras needs to be carefully designed;

[0120] β: The value is very large. The occlusion and undulation of the terrain have a significant impact on the coverage of the camera.

[0121] γ: Specially designed wind turbine layouts may be required to accommodate complex mountainous terrain;

[0122] δ: Very large values, requiring high-performance cameras to overcome the challenges posed by the terrain;

[0123] Number of cameras: The values ​​of k and f(m,n) are greatly increased, resulting in a significantly higher number of cameras than the basic formula to accommodate complex terrain and increased monitoring difficulty.

[0124] IV. Cities or Industrial Areas

[0125] In urban or industrial areas, the layout of wind turbines may be affected by buildings and other man-made structures, so:

[0126] α: large values, because buildings and other structures may cause irregularities in the layout;

[0127] β: The value is large, and the complexity of the urban environment may affect the camera's viewing angle and coverage;

[0128] γ: It may be necessary to consider the impact of buildings on the layout of wind turbines, thereby affecting the coverage requirements of cameras;

[0129] δ: Larger values ​​may require specially designed cameras to adapt to the challenges in urban environments;

[0130] Number of cameras: The number of cameras required must not only take into account the coverage area, but also avoid or cover obstacles such as buildings. Therefore, higher values ​​of k and f(m,n) lead to an increase in the number of cameras required.

[0131] 5. Coastline or Island

[0132] The topographic changes of the coastline or island may affect the layout of the wind turbine and the coverage of the camera. Therefore, in the coastline or island environment, the parameters in the correction function f(m,n) are as follows:

[0133] α: Larger values, because the irregularity of the coastline may affect the uniformity of the layout;

[0134] β: The value is large. Environmental factors on the coastline, such as salt spray and humidity, may affect the performance of the camera.

[0135] γ: It may be necessary to consider the impact of coastline topography on the layout of wind turbines;

[0136] δ: The value is relatively large, requiring a high-performance camera that can adapt to harsh marine environments;

[0137] Number of cameras: Considering environmental factors and the irregularity of the terrain, the values ​​of k and f(m,n) are high, and more cameras are needed to ensure the reliability and stability of the monitoring system.

[0138] According to the above analysis, different installation terrains have a significant impact on camera layout. Therefore, when designing the camera layout optimization algorithm, the impact of terrain factors on the formula parameters must be considered. By reasonably adjusting these parameters, it can be ensured that the algorithm can achieve efficient, economical and reliable camera layout under different terrain conditions, thereby improving the accuracy and efficiency of the wind turbine tilt monitoring system.

[0139] In actual operation, for example, if there are 100 wind turbines in a wind farm, and both m and n are 10, it is simulated as a 10*10 grid map. Therefore, according to the above algorithm, the estimate of 50 cameras can be used as a rough starting point. In fact, more cameras may be needed to ensure full coverage, such as the flat area as the benchmark number, about 50 cameras are needed; the hilly area is about 10-20% of the benchmark number, so about 55 to 60 cameras are needed; the mountainous area is about 30-50% or more of the benchmark number, 65 to 75 cameras; the urban or industrial area is about 20-40% of the benchmark number, so about 60 to 70 cameras are needed; the coastline or island is about 15-30% of the benchmark number, so about 58 to 65 cameras are needed.

[0140] In this embodiment, in order to ensure the stability of the camera in actual operation and reduce the influence of wind direction and wind speed, a strategy for maintaining the stability of the camera is also proposed, which is as follows:

[0141] First, real-time monitoring and data analysis. Real-time monitoring of wind speed and direction is crucial for camera stability. By installing anemometers and wind vanes around the tower, real-time data can be obtained, which helps to analyze the impact of wind on the tower and predict the potential movement of the tower, while providing a scientific basis for adjusting the camera angle.

[0142] Second, automatic adjustment rules. Using advanced automatic adjustment rules, a dynamic response mechanism has been developed based on historical wind data and simulation test results. This mechanism comprehensively considers the dynamic characteristics of wind speed, wind direction and tower structure, and provides precise guidance for camera angle adjustment. By setting key wind speed thresholds, the system can monitor and evaluate wind conditions in real time. Once potential risks are detected, it automatically triggers an early warning and adjusts the camera to a safe observation angle.

[0143] The implementation of this rule ensures that monitoring equipment can respond quickly during extreme wind events, minimize image jitter and data loss, and ensure the continuity and accuracy of monitoring data. At the same time, the close linkage between the system and the main control center provides real-time decision support for operators and enhances the overall safety monitoring capabilities of wind farms.

[0144] Third, image quality monitoring and optimization. Camera jitter caused by wind fluctuations can be effectively controlled by the integrated image stability monitoring system proposed in this invention. The system uses advanced image processing algorithms to monitor and correct camera jitter in real time, and activates optical image stabilization, electronic image stabilization and mechanical shock absorption technology through preset thresholds to ensure image stability and clarity under different wind conditions.

[0145] In a possible embodiment, the present invention further uses a dynamic parameter adjustment function to automatically adjust the exposure time and aperture size of the camera according to changes in wind conditions to optimize image quality. This adaptive adjustment strategy not only improves monitoring efficiency, but also enhances the monitoring safety of wind turbines in complex environments.

[0146] By applying the above strategies, the monitoring system can operate stably under various wind conditions and provide continuous and accurate monitoring data, thereby improving the monitoring efficiency and safety of wind turbines.

[0147] In addition to the layout and installation strategy of the cameras, in this embodiment, in order to ensure that multiple cameras can be interconnected and jointly meet user needs, a Nash equilibrium camera interconnection algorithm is also used to balance the monitoring and tracking range of each camera so that the entire camera interconnection network reaches a stable state.

[0148] Specifically, the core principle of the Nash equilibrium camera interconnection algorithm is based on the concept of Nash equilibrium in game theory. Nash equilibrium is a strategy combination in which any camera will not gain higher utility by changing its strategy alone. In other words, when all cameras operate according to the strategy set by the algorithm, the entire system will reach a stable state, at which time no single camera can improve its monitoring effect by changing its own behavior.

[0149] If a -i =Aa i, which means removing camera C from all targets i The collection of tracked targets. Indicates the tracking target selected by each camera, then the Nash equilibrium is as follows:

[0150]

[0151] in, Represents C i The optimal solution for tracking the target is a specific camera C i ∈C, and Represents the set of all cameras.

[0152] The system utility function is divided into target utility function, overall utility function and camera utility function.

[0153] The target utility function is:

[0154]

[0155] Where A = ∪ i a i , indicating that camera C i All targets to be tracked. The above formula shows that if r 0 Monitor target T i Number of cameras If it is greater than 0, the target T i The utility function is 1 if , otherwise it is 0.

[0156] The overall utility function can be derived from the target utility function:

[0157]

[0158] The camera utility function is:

[0159]

[0160] It can be proved that the utility function proposed above can reach Nash equilibrium.

[0161] In this embodiment, in step S3, the real-time data of the target tower obtained by a camera placed at the center of the top of the adjacent tower is used as initialization data; after the initialization data is set, the target tower is photographed in real time by the camera to obtain and save the real-time image of the target tower.

[0162] Specifically, Figure 3 As shown, place the industrial camera at O 2Adjust the camera to make the image level, set the tower angle and position information at this time as initialization data, and capture and save the wind turbine tower image in real time.

[0163] In this embodiment, in step S4, the real-time image of the target tower taken by the camera is transmitted to the image processing module of the camera Internet network; the real-time image of the target tower and the projection line laser image are processed separately by the image processing module to extract the edge line of the target tower and the laser center line.

[0164] Specifically, the original image of the wind turbine tower is a color RGB image. Since the color feature information of the wind turbine tower is not needed to extract the edge line of the wind turbine tower, the image is grayed to reduce the amount of processing operations. After multiple comparative experiments, in order to completely extract the edge of the wind turbine tower, the weighted average method and the maximum value method were finally used to process the grayed images, and the subsequent processing was performed.

[0165] The weighted average method performs weighted average of each channel according to the importance of the three channels of the color image, and the result is used as the grayscale value of the grayscale image. The calculation formula is as follows:

[0166] Gray(x, y) = C R R(x,y)+C G G(x,y)+C B B(x,y)

[0167] In the formula, the coefficient C R , C G , C B Represent the weights of the three channels of the image, all greater than zero, and C R +C G +C B =1; (x, y) is the coordinate of the pixel point in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel points in the R, G and B channels respectively; the weights are usually selected according to the sensitivity of the human eye to the three colors of red, green and blue, and C is selected. R =0.299, C G =0.587, C B =0.114.

[0168] The maximum value method uses the maximum value of the three channels of the color image as the grayscale value. The calculation formula is as follows:

[0169] Gray(x,y)=max(R(x,y)+max(G(x,y)+B(x,y)))

[0170] Where (x, y) is the coordinate of the pixel in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel in the R, G and B channels respectively.

[0171] After the image is grayed out, it is necessary to filter the image. In the process of the camera collecting wind turbine tower images and converting them digitally, more Gaussian noise will be mixed in, and it will be affected by changes in lighting and damage to the surface of the wind turbine tower, which increases the difficulty of image processing. Use Gaussian filtering to reduce the noise of the grayscale image, and then perform convolution on the image. The calculation formula for the two-dimensional zero mean and standard deviation of Gaussian function is as follows:

[0172]

[0173] Where x and y are the coordinates of the image pixel points; σ is the standard deviation. The larger the σ value is, the smaller the difference between the different coefficient values ​​in the generated Gaussian template is, and the more obvious the smoothing effect on the image is.

[0174] In this embodiment, in step S5, a three-dimensional coordinate system is established according to the center line of the target tower, the wind turbine and the wind turbine impeller; in the three-dimensional coordinate system, the inclination angle and settlement data of the target tower are calculated by extracting the edge line of the target tower and the laser center straight line;

[0175] Among them, the center line of the target tower is the z-axis, the bottom to the top of the tower is the positive direction of the z-axis, the interception point of the bottom of the tower on the z-axis is the coordinate origin, the direction of the wind turbine impeller is the negative direction of the y-axis, and the right side of the front image of the wind turbine is the positive direction of the x-axis, and a three-dimensional coordinate system is established.

[0176] In this embodiment, in step S6, the inclination angle of the target tower is compared and calculated with the inclination angle of the reference wind turbine. If the difference between the two exceeds a preset safety threshold, an alarm is issued; the settlement data of the target tower is compared with the initialization data of the target tower. If a high settlement occurs, an alarm is issued.

[0177] Specifically, in order to ensure the accuracy and reliability of the monitoring results, a group of wind turbines are selected as dynamic reference benchmarks, and their tilt data are updated regularly through manual measurement. These data are used as reference points for subsequent image analysis. The calculation formula for the tilt angle of the reference wind turbine is:

[0178]

[0179] In the formula, θ ref is the inclination angle of the reference fan; h top and h bottom are the heights of the top and bottom of the wind turbine tower respectively, and d is the distance from the bottom to the top of the tower.

[0180] The camera uses built-in sensors (such as gyroscopes) to achieve self-calibration, real-time monitoring and automatic adjustment of posture. The adjustment of the camera tilt angle can be expressed by the following formula:

[0181] θ adj =θ current -θ ref

[0182] In the formula, θ current is the current camera tilt angle, θ adj is the adjusted tilt angle.

[0183] Use image preprocessing technology (such as Gaussian filtering) to enhance feature point recognition. Assume that the standard deviation of the Gaussian filter in the image is σ, then the filtered image I filtered It can be calculated by the following formula:

[0184]

[0185] Where I(x,y) is the original image, I filtered (x,y) is the filtered image.

[0186] The current image is registered with the reference image of the wind turbine by a feature point matching algorithm (such as a feature matching algorithm). Feature point matching can be achieved by minimizing the error between feature points. Let the error function be E:

[0187]

[0188] In the formula, p i is the feature point in the reference image, q i is the corresponding feature point in the current image, and N is the number of feature points.

[0189] The system intelligently identifies the tower's tilt changes through real-time image analysis and compares it with the dynamic reference. Assume that the current tower tilt angle is θ current , when |θ current -θ ref |>∈, an alarm is triggered, where ∈ is the preset safety threshold.

[0190] Through robustness testing and environmental adaptability analysis, the algorithm parameters are optimized to ensure the stability and adaptability of the algorithm under different environmental conditions. Assuming that the impact of environmental changes on algorithm performance is δ, the robustness of the algorithm can be improved by minimizing δ.

[0191] The above algorithm design not only improves the automation level of wind turbine tower tilt detection, but also ensures the accuracy and reliability of monitoring results. It has significant safety and economic benefit potential and is suitable for the monitoring needs of the wind power industry.

[0192] In summary, the present invention measures the height of the target tower and records the actual height of the target tower obtained by measurement; places a camera at the center of the top of an adjacent tower according to the actual height of the target tower and the surrounding environment to ensure that the target tower can be imaged by a set number of adjacent towers; uses the real-time data of the target tower acquired by the camera placed at the center of the top of the adjacent tower as initialization data; after the initialization data is set, the target tower is photographed in real time by the camera to acquire and save the real-time image of the target tower; the real-time image of the target tower photographed by the camera is transmitted to the image processing module of the camera Internet network; and the image processing module is used to process the image of the target tower. The module processes the real-time image and the projection line laser image of the target tower respectively, extracts the edge line and the laser center line of the target tower; establishes a three-dimensional coordinate system according to the center line of the target tower, the fan and the wind turbine impeller; in the three-dimensional coordinate system, calculates the inclination angle and settlement data of the target tower according to the extracted edge line and the laser center line of the target tower; compares and calculates the inclination angle of the target tower with the inclination angle of the reference fan, and if the difference between the two exceeds the preset safety threshold, an alarm is issued; compares the settlement data of the target tower with the initialization data of the target tower, and if a high settlement occurs, an alarm is issued. The present invention can realize all-round and multi-angle monitoring of wind turbine towers by using a multi-camera network. Through the collaborative work between multiple cameras, the inclination of the wind turbine tower can be more comprehensively understood, and the reliability and accuracy of the detection can be improved. Through advanced image acquisition equipment, the original image processing can capture high-definition images of the wind turbine tower, thereby retaining more detailed information. The camera network is used to detect the inclination of the wind turbine tower, realizing non-contact monitoring. This method does not require the installation of sensors or other equipment on the wind turbine tower, which reduces interference with wind turbine equipment and also reduces installation and maintenance costs. Combining computer vision and artificial intelligence technology, the images captured by the camera can be intelligently analyzed to automatically identify and determine the degree of inclination of the wind turbine tower. This intelligent analysis improves the accuracy and efficiency of detection and reduces the need for manual intervention. Through the camera network, the inclination of the wind turbine tower can be displayed to managers or relevant personnel in real time in the form of images or videos. This visual display helps to intuitively understand the operating status of the wind turbine tower, making it easier to find problems in a timely manner and take appropriate measures.

[0193] It should be noted that the method of the embodiment of the present disclosure can be performed by a single device, such as a computer or a server. The method of the present embodiment can also be applied in a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present disclosure, and the multiple devices will interact with each other to complete the described method.

[0194] It should be noted that the above describes some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0195] Example 2

[0196] See also Figure 5 Embodiment 2 of the present invention further provides a tower inclination monitoring device based on a camera Internet network, comprising:

[0197] The tower actual height measurement module 001 is used to measure the height of the target tower and record the measured actual height of the target tower;

[0198] The camera layout placement module 002 is used to place cameras at the center of the top of adjacent towers according to the actual height of the target tower and the surrounding environment, so as to ensure that the target tower can be imaged by a set number of adjacent towers;

[0199] The real-time image acquisition module 003 is used to use the real-time data of the target tower obtained by the camera placed at the center of the top of the adjacent tower as initialization data; after the initialization data is set, the target tower is photographed in real time by the camera to obtain and save the real-time image of the target tower;

[0200] The image data processing module 004 is used to transmit the real-time image of the target tower taken by the camera to the image processing module of the camera Internet network; the real-time image of the target tower and the projection line laser image are processed by the image processing module respectively to extract the edge line of the target tower and the laser center line;

[0201] The tower real-time tilt angle and settlement data acquisition module 005 is used to establish a three-dimensional coordinate system according to the center line of the target tower, the wind turbine and the wind turbine impeller; in the three-dimensional coordinate system, the tilt angle and settlement data of the target tower are calculated by extracting the edge line of the target tower and the laser center straight line;

[0202] The data comparison alarm module 006 is used to compare and calculate the inclination angle of the target tower with the inclination angle of the reference wind turbine. If the difference between the two exceeds a preset safety threshold, an alarm reminder is issued; the settlement data of the target tower is compared with the initialization data of the target tower. If a high settlement occurs, an alarm reminder is issued.

[0203] In this embodiment, in the camera layout placement module 002, in order to ensure that the target tower can be imaged by a set number of adjacent towers, a camera is installed on the tower using a camera installation algorithm; the camera is calculated according to the algorithm as follows:

[0204]

[0205] Where C is the total number of cameras required; m and n represent the number of wind turbines in the horizontal and vertical directions, respectively; k is the number of additional cameras due to terrain, obstacles or specified environmental factors; f(m,n) is a correction function used to take into account the irregularity of the layout and other influencing factors;

[0206] In the camera layout placement module, the Nash equilibrium camera interconnection algorithm is also used to enable the cameras in the camera interconnection network to maintain stable operation; the expression of the Nash equilibrium camera interconnection algorithm is:

[0207]

[0208] in, Represents C i The optimal solution for tracking the target is a specific camera C i ∈C, and Represents the set of all cameras.

[0209] In this embodiment, in the data comparison alarm module 006, in the process of comparing and calculating the inclination angle of the target tower with the inclination angle of the reference wind turbine, the calculation formula of the inclination angle of the reference wind turbine is:

[0210]

[0211] In the formula, θ ref is the inclination angle of the reference fan; h top and h bottomare the heights of the top and bottom of the wind turbine tower respectively, and d is the distance from the bottom to the top of the tower.

[0212] In this embodiment, in the image data processing module 004, in the process of respectively processing the real-time image and the projection line laser image of the target tower by the image processing module, extracting the edge line of the target tower and the laser center straight line, the real-time image of the target tower is subjected to image grayscale processing by weighted average method and maximum value method;

[0213] The calculation formula of the weighted average method is:

[0214] Gray(x, y) = C R R(x,y)+C G G(x,y)+C B B(x,y)

[0215] In the formula, the coefficient C R , C G , C B Represent the weights of the three channels of the image, all greater than zero, and C R +C G +C B =1; (x, y) is the coordinate of the pixel point in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel points in the R, G and B channels respectively;

[0216] The calculation formula of the maximum value method is:

[0217] Gray(x,y)=max(R(x,y)+max(G(x,y)+B(x,y)))

[0218] Where (x, y) is the coordinate of the pixel point in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel points in the R, G and B channels respectively;

[0219] After the real-time image of the target tower is grayed, the real-time image of the target tower that has been grayed is subjected to noise reduction processing by Gaussian filtering. The calculation formula of the Gaussian function with two-dimensional zero mean and standard deviation σ is:

[0220]

[0221] Where x and y are the coordinates of the image pixel points; σ is the standard deviation. The larger the σ value is, the smaller the difference between the different coefficient values ​​in the generated Gaussian template is, and the more obvious the smoothing effect on the image is.

[0222] It should be noted that the information interaction, execution process and other contents between the modules of the above-mentioned system are based on the same concept as the method embodiment in Example 1 of the present application, and the technical effects they bring are the same as those of the method embodiment of the present application. For specific contents, please refer to the description in the method embodiment shown above in the present application, and will not be repeated here.

[0223] Example 3

[0224] Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium, in which a program code of a tower inclination monitoring method based on a camera Internet network is stored. The program code includes instructions for executing embodiment 1 or any possible implementation method of a tower inclination monitoring method based on a camera Internet network.

[0225] The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0226] Example 4

[0227] Embodiment 4 of the present invention provides an electronic device, including: a memory and a processor;

[0228] The processor and the memory communicate with each other via a bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute a tower inclination monitoring method based on a camera Internet network in Example 1 or any possible implementation thereof.

[0229] Specifically, the processor can be implemented by hardware or by software. When implemented by hardware, the processor can be a logic circuit, an integrated circuit, etc.; when implemented by software, the processor can be a general-purpose processor implemented by reading software codes stored in a memory. The memory can be integrated in the processor or can be located outside the processor and exist independently.

[0230] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable systems. The computer instructions can be stored in a computer-readable storage medium, or transmitted from a computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center.

[0231] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing system, they can be concentrated on a single computing system, or distributed on a network composed of multiple computing systems, and optionally, they can be implemented by a program code executable by a computing system, so that they can be stored in a storage system and executed by the computing system, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0232] Although the present invention has been described in detail above by general description and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made on the basis of the present invention. Therefore, these modifications or improvements made on the basis of not departing from the spirit of the present invention all belong to the scope of protection claimed by the present invention.

Claims

1. A tower inclination monitoring method based on a camera Internet network, characterized in that: include: Measuring the height of a target tower and recording the measured actual height of the target tower; According to the actual height of the target tower and the surrounding environment, a camera is placed at the center of the top of the adjacent tower to ensure that the target tower is imaged by a set number of adjacent towers; The real-time data of the target tower obtained by a camera placed at the center of the top of the adjacent tower is used as initialization data; after the initialization data is set, the target tower is photographed in real time by the camera to obtain and save the real-time image of the target tower; The real-time image of the target tower taken by the camera is transmitted to the image processing module of the camera Internet network; the real-time image of the target tower and the projection line laser image are processed by the image processing module respectively to extract the edge line of the target tower and the laser center line; A three-dimensional coordinate system is established according to the center line of the target tower, the wind turbine and the wind turbine impeller; in the three-dimensional coordinate system, the inclination angle and settlement data of the target tower are calculated based on the edge line of the target tower and the laser center line extracted; Compare and calculate the inclination angle of the target tower with the inclination angle of the reference wind turbine, and if the difference between the two exceeds a preset safety threshold, issue a tilt alarm reminder; compare the settlement data of the target tower with the initialization data of the target tower, and if a high settlement occurs, issue a settlement alarm reminder; To ensure that the target tower can be imaged by a set number of adjacent towers, a camera is installed on the tower using a camera installation algorithm; the calculation formula of the camera installation algorithm is: Where C is the total number of cameras required; m and n represent the number of wind turbines in the horizontal and vertical directions, respectively; k is the number of additional cameras due to terrain, obstacles or specified environmental factors; f(m,n) is a correction function used to take into account the irregularity of the layout and other influencing factors; The correction function f(m,n) is defined as: f(m,n)=α*g(m,n)+β*h(m,n)+γ*s+δ*p+∈*q Where g(m, n) is a function related to layout irregularity; h(m, n) is a function related to environmental factors; s is the average size of a wind turbine; p represents the average performance index of the camera; q represents the required level of monitoring accuracy; The Nash equilibrium camera interconnection algorithm is used to keep the cameras in the camera interconnection network running stably. The expression of the Nash equilibrium camera interconnection algorithm is: in, Represents C i The optimal solution for tracking the target is a specific camera C i ∈C, and Represents the set of all cameras; A group of wind turbines are selected as dynamic reference benchmarks, and their tilt data are updated regularly through manual measurement; In the process of comparing and calculating the inclination angle of the target tower with the inclination angle of the reference wind turbine, the calculation formula of the inclination angle of the reference wind turbine is: In the formula, θ ref is the inclination angle of the reference fan; h top and h bottom are the heights of the top and bottom of the wind turbine tower, respectively, and d is the distance from the bottom to the top of the tower; The camera uses built-in sensors to achieve self-calibration, real-time monitoring and automatic adjustment of posture. The adjustment of the camera tilt angle is expressed by the following formula: i adj =θ current -θ ref In the formula, θ current is the current camera tilt angle, θ adj is the adjusted tilt angle; Image preprocessing technology is used to enhance feature point recognition. Assuming the standard deviation of the Gaussian filter in the image is σ, the filtered image I filtered Calculated by the following formula: Where I(x, y) is the original image, I filtered (x, y) is the filtered image; The current image is registered with the reference image of the wind turbine through the feature point matching algorithm. The feature point matching is achieved by minimizing the error between feature points. Let the error function be E: In the formula, p i is the feature point in the reference image, q i is the corresponding feature point in the current image, and N is the number of feature points; The system intelligently identifies the inclination change of the tower through real-time image analysis and compares it with the dynamic reference benchmark; assuming that the current inclination angle of the tower is θ current , when |θ current -θ ref |>∈, an alarm is triggered, where ∈ is the preset safety threshold.

2. The tower inclination monitoring method based on camera Internet network according to claim 1 is characterized in that: In the process of respectively processing the real-time image and the projection line laser image of the target tower by the image processing module and extracting the edge line of the target tower and the laser center straight line, the real-time image of the target tower is grayed by the weighted average method and the maximum value method; The calculation formula of the weighted average method is: Gray(x,y)=C R R(x,y)+C G G(x,y)+C B B(x,y) In the formula, the coefficient C R , C G , C B Represent the weights of the three channels of the image, all greater than zero, and C R +C G +C B =1; (x, y) is the coordinate of the pixel point in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel points in the R, G and B channels respectively; The calculation formula of the maximum value method is: Gray(x,y)=max(R(x,y)+max(G(x,y)+B(x,y))) Where (x, y) is the coordinate of the pixel in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel in the R, G and B channels respectively.

3. The tower inclination monitoring method based on camera Internet network according to claim 2 is characterized in that: In the process of respectively processing the real-time image and the projection line laser image of the target tower by the image processing module and extracting the edge line of the target tower and the laser center line, the real-time image of the target tower that has completed the grayscale processing is subjected to noise reduction processing by Gaussian filtering. The calculation formula of the Gaussian function with two-dimensional zero mean and standard deviation σ is: Where x and y are the coordinates of the image pixel points; σ is the standard deviation. The larger the σ value is, the smaller the difference between the different coefficient values ​​in the generated Gaussian template is, and the more obvious the smoothing effect on the image is.

4. A tower inclination monitoring device based on a camera Internet network, using a tower inclination monitoring method based on a camera Internet network as claimed in any one of claims 1 to 3, characterized in that: include: The actual tower height measurement module is used to measure the height of the target tower and record the measured actual height of the target tower; A camera layout placement module is used to place cameras at the center of the top of adjacent towers according to the actual height of the target tower and the surrounding environment, so as to ensure that the target tower can be imaged by a set number of adjacent towers; A real-time image acquisition module is used to use the real-time data of the target tower obtained by a camera placed at the center of the top of the adjacent tower as initialization data; after the initialization data is set, the target tower is photographed in real time by the camera to obtain and save the real-time image of the target tower; An image data processing module is used to transmit the real-time image of the target tower taken by the camera to the image processing module of the camera Internet network; the real-time image of the target tower and the projection line laser image are processed by the image processing module respectively to extract the edge line of the target tower and the laser center line; A tower real-time tilt angle and settlement data acquisition module is used to establish a three-dimensional coordinate system according to the center line of the target tower, the fan and the wind turbine impeller; in the three-dimensional coordinate system, the tilt angle and settlement data of the target tower are calculated by extracting the edge line of the target tower and the laser center straight line; A data comparison alarm module is used to compare and calculate the inclination angle of the target tower with the inclination angle of the reference wind turbine, and issue an alarm if the difference between the two exceeds a preset safety threshold; compare the settlement data of the target tower with the initialization data of the target tower, and issue an alarm if a high settlement occurs; In the camera layout placement module, in order to ensure that the target tower can be imaged by a set number of adjacent towers, a camera is installed on the tower using a camera installation algorithm; the calculation formula of the camera installation algorithm is: Where C is the total number of cameras required; m and n represent the number of wind turbines in the horizontal and vertical directions, respectively; k is the number of additional cameras due to terrain, obstacles or specified environmental factors; f(m,n) is a correction function used to take into account the irregularity of the layout and other influencing factors; The correction function f(m,n) is defined as: f(m,n)=α*g(m,n)+β*h(m,n)+γ*s+δ*p+∈*q Where g(m,n) is a function related to layout irregularity; h(m,n) is a function related to environmental factors; s is the average size of a wind turbine; p represents the average performance index of the camera; q represents the required level of monitoring accuracy; In the camera layout placement module, the Nash equilibrium camera interconnection algorithm is also used to enable the cameras in the camera interconnection network to maintain stable operation; the expression of the Nash equilibrium camera interconnection algorithm is: in, Represents C i The optimal solution for tracking the target is a specific camera C i ∈C, and Represents the set of all cameras; A group of wind turbines are selected as dynamic reference benchmarks, and their tilt data are updated regularly through manual measurement; In the data comparison alarm module, in the process of comparing and calculating the inclination angle of the target tower with the inclination angle of the reference wind turbine, the calculation formula of the inclination angle of the reference wind turbine is: In the formula, θ ref is the inclination angle of the reference fan; h top and h bottom are the heights of the top and bottom of the wind turbine tower, respectively, and d is the distance from the bottom to the top of the tower; The camera uses built-in sensors to achieve self-calibration, real-time monitoring and automatic adjustment of posture. The adjustment of the camera tilt angle is expressed by the following formula: i adj =θ current -θ ref In the formula, θ current is the current camera tilt angle, θ adj is the adjusted tilt angle; Image preprocessing technology is used to enhance feature point recognition. Assuming the standard deviation of the Gaussian filter in the image is σ, the filtered image I filtered Calculated by the following formula: Where I(x,y) is the original image, I filtered (x,y) is the filtered image; The current image is registered with the reference image of the wind turbine through the feature point matching algorithm. The feature point matching is achieved by minimizing the error between feature points. Let the error function be E: In the formula, p i is the feature point in the reference image, q i is the corresponding feature point in the current image, and N is the number of feature points; The system intelligently identifies the inclination change of the tower through real-time image analysis and compares it with the dynamic reference benchmark; assuming that the current inclination angle of the tower is θ current , when |θ current -θ ref |>∈, an alarm is triggered, where ∈ is the preset safety threshold.

5. The tower inclination monitoring device based on camera internet network according to claim 4 is characterized in that: In the image data processing module, the real-time image and the projection line laser image of the target tower are processed respectively by the image processing module, and in the process of extracting the edge line of the target tower and the laser center line, the real-time image of the target tower is grayed by the weighted average method and the maximum value method; The calculation formula of the weighted average method is: Gray(x,y)=C R R(x,y)+C G G(x,y)+C B B(x,y) In the formula, the coefficient C R , C G , C B Represent the weights of the three channels of the image, all greater than zero, and C R +C G +C B =1; (x, y) is the coordinate of the pixel point in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel points in the R, G and B channels respectively; The calculation formula of the maximum value method is: Gray(x,y)=max(R(x,y)+max(G(x,y)+B(x,y))) Where (x, y) is the coordinate of the pixel point in the image; R(x, y), G(x, y) and B(x, y) are the grayscale values ​​of the pixel points in the R, G and B channels respectively; After the real-time image of the target tower is grayed, the real-time image of the target tower that has been grayed is subjected to noise reduction processing by Gaussian filtering. The calculation formula of the Gaussian function with two-dimensional zero mean and standard deviation σ is: Where x and y are the coordinates of the image pixel points; σ is the standard deviation. The larger the σ value is, the smaller the difference between the different coefficient values ​​in the generated Gaussian template is, and the more obvious the smoothing effect on the image is.

Images