A method and system for non-contact monitoring of cable force of photovoltaic flexible support based on video vibration

Abstract

The present application belongs to the technical field of engineering structure health monitoring, and discloses a photovoltaic flexible support cable force non-contact monitoring method and system based on video vibration. First, a cable video sequence is acquired, and a dynamic sensitive area is adaptively divided along the cable image. The time sequence average value of the pixel intensity in each area is calculated, the two-dimensional image information is converted into a one-dimensional vibration time history signal, and the light drift and image noise are suppressed through difference and wavelet denoising. After filtering the signal of each dynamic sensitive area, the initial frequency set is formed by preliminarily extracting the multi-order dominant frequency estimate value. The frequency observations from all areas are fused and analyzed, and the real frequency cluster is identified. The weighted centroid frequency of the cluster is used as the multi-order vibration frequency. Finally, the multi-order frequency is substituted into the model, and the real-time cable force value is output. The whole process of the method does not need to contact the measured cable, and provides an innovative and reliable solution for long-term online safety monitoring of the flexible support cable force in complex environments.

Description

Technical Field

[0001] This invention belongs to the field of engineering structure health monitoring technology, specifically relating to a non-contact monitoring method and system for the cable force of photovoltaic flexible support based on video vibration. Background Technology

[0002] The cable force of flexible supports is a core indicator of their stress state and structural safety, and accurate and reliable monitoring of it is crucial to ensuring the safe operation of engineering structures. Currently, traditional cable force measurement methods mainly rely on contact sensors, which have several limitations. One is the pressure sensor method, which measures force by directly installing force sensors at the anchorages. While this method offers high accuracy, it requires pre-installation during construction, resulting in high costs and making it difficult to implement on existing structures. Another method is the vibration frequency method based on string vibration theory. This method uses accelerometers installed on the cable to pick up vibration signals and analyze their frequencies to inversely determine the cable force; this technology is relatively mature. However, this method requires the installation and maintenance of physical sensors on the cable. For high-altitude, long-span cable structures or electrified environments, this presents problems such as installation difficulties, maintenance inconvenience, and safety hazards. Furthermore, constructing a long-term, large-scale monitoring system involves complex wiring and high overall costs.

[0003] In summary, existing cable force monitoring technologies are either limited by pre-embedding requirements during the construction phase or constrained by the installation and maintenance difficulties of physical sensors, making it difficult to simultaneously meet the needs for high precision, low cost, ease of implementation, and long-term stable monitoring. Summary of the Invention

[0004] The purpose of this invention is to solve the problems of existing cable force monitoring relying on contact sensors, which are difficult to install and maintain and have insufficient adaptability, and to provide a non-contact monitoring method and system for photovoltaic flexible support cable force based on video vibration.

[0005] To achieve the above objectives, the present invention employs the following technical solution: This invention proposes a non-contact monitoring method for the cable force of a photovoltaic flexible support based on video vibration, comprising the following steps: A preprocessed video frame sequence is obtained, and the preprocessed video frame sequence is projected along the cable body into the image to obtain multiple dynamic regions of interest. Each dynamic region of interest is processed to generate a pixel intensity-time history signal, which serves as the original vibration signal characterizing the local lateral vibration of the cable segment. The original vibration signal is filtered and subjected to variational mode decomposition to obtain modal components. The spectrum of each modal component is analyzed, and the estimated values ​​of the first K dominant vibration frequencies are identified and extracted to form an initial frequency set. The initial frequency set is summarized to form a high-dimensional frequency observation dataset. The high-dimensional frequency observation dataset is analyzed to obtain the overall, spatially consistent true vibration frequencies of each order of the photovoltaic flexible support cable structure. A refined cable force calculation model is established, and the actual vibration frequencies of each order and their corresponding estimated orders are substituted into the model for inversion and solution, outputting real-time cable force values, thereby realizing non-contact monitoring of the cable force of photovoltaic flexible supports.

[0006] Preferably, the process of projecting the preprocessed video frame sequence along the cable body into the image to obtain multiple dynamic regions of interest specifically involves: Along the centerline of the cable body, the cable body image region is divided into N consecutive dynamic regions of interest, each with a width of... Adaptive adjustment:

[0007] in, LocalWidth (i) For the cable body in the first i The width of the imaging pixels at each location α This is a scaling factor; for curved cable segments, the ROI is divided along the tangent to the cable's centerline to ensure that the axis of the dynamically interested region is perpendicular to the direction of the cable's local vibration.

[0008] Preferably, the step of processing each dynamic region of interest to generate a pixel intensity-time history signal as the original vibration signal characterizing the local lateral vibration of the cable segment specifically involves: For the j The first frame of the image i For each dynamic region of interest, calculate the average intensity of all pixels within that region in the green channel. I i,j :

[0009] in, M This represents the total number of pixels within the ROI. G m ( i , j ) represents the intensity value of the m-th pixel in the G channel within the dynamically defined region of interest; For all video frames j =1,2,…, J Repeat this process to obtain the first... i Pixel intensity-time history signal sequence of a dynamic region of interest , For the first i The average pixel intensity of each ROI in the first frame image. For the firsti The average pixel intensity of each ROI in the last frame of the image; For each pixel intensity-time history signal sequence First-order difference and wavelet threshold denoising are performed to obtain pixel intensity-time history signals.

[0010] Preferably, the step of filtering and performing variational mode decomposition on the original vibration signal to obtain modal components, analyzing the spectrum of each modal component, identifying and extracting the estimated values ​​of the first K dominant vibration frequencies, and forming an initial frequency set specifically involves: Bandpass filtering is performed on the original vibration signal of each dynamic region of interest to obtain the expected frequency range of the signal. f low , f high ];in, f low To be the minimum expected frequency, f high This represents the maximum expected frequency. For each filtered signal x i ( t The variational mode decomposition algorithm is applied to adaptively decompose the signal into... K Individual eigenmode functions u k ( t ):

[0011] in, It is the first One modal component, It is its center frequency. It is the Dirac function. The convolution is represented by a quadratic penalty term and Lagrange multipliers, and the solution is iteratively obtained using the alternating direction multiplier method until the convergence condition is met; the number of modes... K Determined automatically by observing the spectrum or using the center frequency method; For each modal component obtained from the decomposition Perform a Fast Fourier Transform to obtain its spectrum. Identify the spectrum The frequency corresponding to the highest peak value of the medium amplitude The frequency of all modes extracted from all dynamic regions of interest. The data are then compiled to form an initial frequency observation set. ,in ; For the first frequency observation, This is the second frequency observation. For the first Each frequency observation.

[0012] Preferably, the aggregated initial frequency set constitutes a high-dimensional frequency observation dataset. Analysis of the high-dimensional frequency observation dataset yields the spatially consistent true vibration frequencies of each order for the entire photovoltaic flexible support cable structure. Specifically: Initial frequency observation set Treating the dataset as a high-dimensional observation dataset, principal component analysis is performed to calculate the eigenvalues ​​and eigenvectors of the dataset's covariance matrix. The original data is then projected onto the space corresponding to the principal components to obtain the dataset. ; For dataset A density-based spatial clustering algorithm is applied to divide data points into core points, boundary points, and noise points. A set of density-connected core points and boundary points constitutes a cluster, corresponding to different orders of true vibration frequencies. Points marked as noise are outliers and are removed. The spatial clustering algorithm identifies the first 1 effective cluster Calculate the weighted centroid of all member frequency values ​​as the true frequency of that order. :

[0013] Among them, weight This involves determining the peak amplitude or signal-to-noise ratio of the modal component corresponding to that frequency, ultimately resulting in an ordered, spatially consistent set of multi-order true frequencies. ,in, , The first vibration frequency of the cable is... The second-order vibration frequency, It is the Mth order vibration frequency.

[0014] Preferably, the establishment of the refined cable force calculation model specifically includes:

[0015] in, F n Indicates the first n First vibration frequency; n Indicates the vibration order; L e Indicates the effective vibration length; T Indicates the force to be claimed; ρ e It represents the equivalent linear density, including the mass of the cable itself and the attached material; The comprehensive correction function is expressed as follows:

[0016] This indicates a correction term for bending stiffness; This indicates the temperature correction term. This indicates the temperature coefficient of the elastic modulus of the cable. Indicates real-time temperature; Indicates reference temperature; This represents the boundary stiffness correction factor, obtained through modal analysis fitting of the finite element model or on-site calibration, and relates to the equivalent boundary stiffness. K eq A function of order n.

[0017] Preferably, the step of substituting the actual vibration frequencies of each order and their corresponding estimated orders into the model for inversion and solving, and outputting the real-time cable force value, specifically involves: The true vibration frequencies of each order and its corresponding estimation order Substitute the values ​​into the refined cable force calculation model used for photovoltaic flexible supports; The cable forces are solved using the weighted least squares method. The optimization problem is expressed as:

[0018] Among them, weight With the The confidence level of order frequency identification is proportional to that of clusters. The density or number of interior points determines the cable force; by iteratively solving the above nonlinear least squares problem, the real-time cable force value is output. .

[0019] This invention proposes a non-contact monitoring system for the cable force of a photovoltaic flexible support based on video vibration, comprising: The vibration signal extraction module is used to acquire a preprocessed video frame sequence, project the preprocessed video frame sequence along the cable body into the image, acquire multiple dynamic regions of interest, process each dynamic region of interest to generate a pixel intensity-time history signal, which serves as the original vibration signal characterizing the local lateral vibration of the cable segment. The frequency identification and extraction module is used to filter and perform variational mode decomposition on the original vibration signal, obtain modal components, analyze the spectrum of each modal component, identify and extract the estimated values ​​of the first K dominant vibration frequencies, and form an initial frequency set. The vibration frequency fusion module is used to summarize the initial frequency set to form a high-dimensional frequency observation dataset, analyze the high-dimensional frequency observation dataset, and obtain the overall, spatially consistent true vibration frequencies of each order of the photovoltaic flexible support cable structure. The support cable force acquisition module is used to establish a refined cable force calculation model, substitute the actual vibration frequencies of each order and their corresponding estimated orders into the model for inversion and solution, and output the real-time cable force value to realize non-contact monitoring of the cable force of photovoltaic flexible support.

[0020] A terminal device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of a non-contact monitoring method for the cable force of a photovoltaic flexible support based on video vibration.

[0021] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of a non-contact monitoring method for the cable force of a photovoltaic flexible support based on video vibration.

[0022] Compared with the prior art, the present invention has the following beneficial effects: This invention proposes a non-contact monitoring method for the cable force of a photovoltaic flexible support based on video vibration. First, a clear video sequence of the cable under environmental wind-induced and operational micro-vibrations is acquired. Then, the dynamic interest region is adaptively divided along the cable image. By calculating the temporal average of pixel intensity within each dynamic interest region, the two-dimensional image information is converted into a one-dimensional vibration time-history signal. Differential and wavelet denoising are used to suppress illumination drift and image noise, achieving a robust conversion from spatial motion to temporal signal. Next, the signal from each dynamic interest region is filtered, and variational mode decomposition is used to adaptively separate multiple eigenmodes. Preliminary estimates of the multi-order dominant frequencies reflected at each location are extracted, forming an initial frequency set. To eliminate outliers caused by local noise and non-structural vibration, a strategy combining principal component analysis and density-based spatial clustering is adopted to fuse and analyze frequency observations from all dynamic interest regions. This identifies spatially consistent and dense true frequency clusters, and the weighted centroid frequency of the cluster is used as the reliable multi-order vibration frequency of the cable structure as a whole. Finally, the aforementioned multi-order frequencies are substituted into the refined mechanical model established for the photovoltaic flexible support. This model incorporates the effects of bending stiffness, equivalent boundary elastic constraints, and temperature on material properties through a comprehensive correction function system. A weighted least squares method with frequency identification confidence as the weight is used for inversion solution, ultimately outputting high-precision real-time cable force values. This method requires no contact with the cable under test throughout the entire process, overcoming the bottlenecks of traditional sensors in terms of installation, maintenance, and cost. Through the deep integration of visual measurement with advanced signal processing, data fusion, and physical modeling technologies, it provides an innovative and reliable solution for long-term online safety monitoring of flexible support cable forces in complex environments.

[0023] This invention proposes a non-contact monitoring system for the cable force of flexible photovoltaic supports based on video vibration. By dividing the system into a vibration signal extraction module, a frequency identification and extraction module, a vibration frequency fusion module, and a support cable force acquisition module, it obtains real-time cable force values, achieving non-contact monitoring of the cable force of flexible photovoltaic supports. The modular design ensures that each module is independent, facilitating unified management of all modules. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of the non-contact monitoring method for the cable force of a photovoltaic flexible support based on video vibration, as described in this invention.

[0026] Figure 2 This is a diagram of the non-contact monitoring system for the cable force of a photovoltaic flexible support based on video vibration, as described in this invention.

[0027] Figure 3 This is a schematic diagram of the structure of an electronic device according to the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0029] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0030] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0031] The present invention will now be described in further detail with reference to the accompanying drawings: To address the problems existing in current technologies, this invention proposes a novel cable force monitoring method that is sensorless, easy to install, low in maintenance costs, and applicable to complex working conditions. This method aims to overcome the bottlenecks of traditional technologies and ensure the safe operation of engineering structures. The invention proposes a non-contact monitoring method for the cable force of photovoltaic flexible supports based on video vibration, such as... Figure 1 As shown, it includes the following steps: S1. Obtain the preprocessed video frame sequence, project the preprocessed video frame sequence along the cable body into the image, obtain multiple dynamic regions of interest, process each dynamic region of interest to generate a pixel intensity-time history signal, which serves as the original vibration signal characterizing the local lateral vibration of the cable segment. Along the centerline of the cable body, the cable body image region is divided into N consecutive dynamic regions of interest, each with a width of... Adaptive adjustment; for curved cable segments, the division direction of the ROI is along the tangent direction of the cable centerline to ensure that the axis of the dynamically interested region is perpendicular to the local vibration direction of the cable.

[0032] For the j The first frame of the image i For each dynamic region of interest, calculate the average intensity of all pixels within that region in the green channel. I i,j ; in, M This represents the total number of pixels within the ROI. G m ( i , j ) represents the intensity value of the m-th pixel in the G channel within the dynamically defined region of interest; For all video frames j =1,2,…, J Repeat this process to obtain the first... i Pixel intensity-time history signal sequence of a dynamic region of interest ; For the first i The ROIs in the first frame ( j =1) Average pixel intensity in the image, For the first i The ROI in the last frame ( j = J The average pixel intensity in the image; For each pixel intensity-time history signal sequence First-order difference and wavelet threshold denoising are performed to obtain pixel intensity-time history signals.

[0033] S2. Filter and perform variational mode decomposition on the original vibration signal to obtain modal components, analyze the spectrum of each modal component, identify and extract the estimated values ​​of the first K dominant vibration frequencies, and form an initial frequency set. Bandpass filtering is performed on the original vibration signal of each dynamic region of interest to obtain the expected frequency range of the signal. f low , f high ];in, f low To be the minimum expected frequency, f high This represents the maximum expected frequency. For each filtered signal x i ( t The variational mode decomposition algorithm is applied to adaptively decompose the signal into... K Individual eigenmode functions u k ( t ):

[0034] in, It is the first One modal component, It is its center frequency. It is the Dirac function. The convolution is represented by a quadratic penalty term and Lagrange multipliers, and the solution is iteratively obtained using the alternating direction multiplier method until the convergence condition is met; the number of modes... K Determined automatically by observing the spectrum or using the center frequency method; For each modal component obtained from the decomposition Perform a Fast Fourier Transform to obtain its spectrum. Identify the spectrum The frequency corresponding to the highest peak value of the medium amplitude The frequency of all modes extracted from all dynamic regions of interest. The data are then compiled to form an initial frequency observation set. ,in ; For the first frequency observation, This is the second frequency observation. This is the p-th frequency observation.

[0035] S3. Summarize the initial frequency set to form a high-dimensional frequency observation dataset. Analyze the high-dimensional frequency observation dataset to obtain the overall, spatially consistent true vibration frequencies of each order of the photovoltaic flexible support cable structure. Initial frequency observation set Treating the dataset as a high-dimensional observation dataset, principal component analysis is performed to calculate the eigenvalues ​​and eigenvectors of the dataset's covariance matrix. The original data is then projected onto the dimensional space corresponding to the principal components to obtain the dataset. ; For dataset A density-based spatial clustering algorithm is applied to divide data points into core points, boundary points, and noise points. A set of density-connected core points and boundary points constitutes a cluster, corresponding to different orders of true vibration frequencies. Points marked as noise are outliers and are removed. The spatial clustering algorithm identifies the first 1 effective cluster Calculate the weighted centroid of all member frequency values ​​as the true frequency of that order. Ultimately, an ordered, spatially consistent set of multi-order true frequencies is obtained. ,in , The first vibration frequency of the cable is... The second-order vibration frequency, It is the Mth order vibration frequency.

[0036] S4. Establish a refined cable force calculation model, substitute the actual vibration frequencies of each order and their corresponding estimated orders into the model for inversion and solution, output the real-time cable force value, and realize non-contact monitoring of the cable force of photovoltaic flexible support.

[0037] The true vibration frequencies of each order and its corresponding estimation order Substitute the values ​​into the refined cable force calculation model used for photovoltaic flexible supports; The cable forces are solved using the weighted least squares method. The optimization problem is expressed as:

[0038] Among them, weight With the The confidence level of order frequency identification is proportional to that of clusters. The density or number of interior points determines the cable force; by iteratively solving the above nonlinear least squares problem, the real-time cable force value is output. .

[0039] The method is described in detail below: A non-contact monitoring method for cable force of photovoltaic flexible support based on video vibration is proposed. The photovoltaic flexible support is a spatial structure composed of prestressed cables, support rods, and photovoltaic panels. Its cable structure has high-frequency, low-amplitude vibration characteristics and is in a multi-field coupled environment of wind, light, and heat. The method includes the following steps: Step 1, Vibration-oriented video acquisition: Deploy image acquisition equipment at a safe monitoring distance from the target cable segment, adjust the optical zoom and field of view to ensure clear imaging of the target cable segment without artificial targets; acquire a continuous video sequence under environmental excitation, including natural wind load and micro-vibrations caused by photovoltaic panel operation.

[0040] Monitoring point planning: Based on the photovoltaic array layout diagram and structural calculation model, key cable segments are identified, including mid-span, near connection points, and locations where internal forces are estimated to be at their maximum or minimum. While ensuring unobstructed visibility and avoiding permanent obstructions, one or more fixed monitoring points are selected on the ground, adjacent supports, or a dedicated observation platform. A safe monitoring distance (typically 5 to 50 meters) must be maintained between the monitoring points and the target cable segments. This distance must ensure imaging accuracy while preventing equipment from entering the structural fall risk zone and facilitating routine maintenance.

[0041] Equipment Selection and Installation: Industrial-grade high-speed cameras or high-definition network smart cameras with high dynamic range and low-light performance are required. Key parameters must meet the following requirements: resolution no less than 1920×1080 pixels to ensure sufficient pixel coverage of the cable structure in the image at long distances (ideally ≥10 pixels in width). The frame rate should be configurable and no less than twice the estimated highest measured frequency of the target cable segment, taking into account the high-frequency micro-vibration characteristics of the photovoltaic cable; the frame rate is typically set to 50fps to 200fps. For scenarios with frequent wind speed changes, an adaptive frame rate mode can be used. The sensor and lens should be equipped with a large-area sensor and a telephoto zoom lens with optical image stabilization, supporting remote close-up shooting.

[0042] Securely mount the device on a heavy-duty tripod or a dedicated vibration-damping platform to ensure absolute camera stability throughout the entire acquisition cycle, eliminating false displacements caused by device vibration. Power and communication are provided via wired (PoE) or wireless (4G / 5G) connections to the local monitoring network.

[0043] Step 2, Highly Robust Vibration Signal Extraction: For the preprocessed video frame sequence, project along the slender axis of the cable body in the image, and adaptively divide a series of continuous, equal-width or variable-width pixel rows or rectangular strips as dynamic regions of interest (ROIs) to match the bending shape of the cable; for each ROI, generate a one-dimensional pixel intensity-time history signal by calculating the temporal average of the gray values ​​of all pixels or the intensity of a specific color channel within it, as the original signal characterizing the local lateral vibration of the cable segment.

[0044] Step 2.1, Dynamic ROI Adaptive Partitioning.

[0045] Following the direction of the cable body centerline coarsely extracted in step S1, the cable body image region is divided into N consecutive rectangular strips (ROIs). The width of each ROI is... Adaptive adjustment:

[0046] in, LocalWidth (i) For the cable body in the first i The width of the imaging pixels at each location α This is a scaling factor (usually 0.8-1.2) to ensure that the ROI completely encloses the cable body but does not excessively include the background. For curved cable segments, the ROI is divided along the tangent to the cable centerline to ensure that the ROI axis is perpendicular to the direction of local vibration of the cable.

[0047] Step 2.2, Pixel intensity timing signal generation.

[0048] For the j The first frame of the image i For each region of interest (ROI), calculate the average intensity of all pixels within it in the green channel (G channel, as it typically has the highest signal-to-noise ratio and contrast). I i,j :

[0049] in, M This represents the total number of pixels within the ROI. G m ( i , j ) represents the intensity value (0-255) of the m-th pixel in the G channel within the ROI.

[0050] For all video frames j =1,2,…, J Repeat this process to obtain the first... i Pixel intensity-time series of ROIs The fluctuations in the sequence reflect the lateral vibration of the cable at that location, perpendicular to the imaging plane.

[0051] Step 2.3, Signal noise reduction and trend removal.

[0052] For each I i First-order differencing is performed to eliminate any potential slow illumination drift. Wavelet thresholding is then used to preprocess the differencing signal to suppress high-frequency image noise.

[0053] Step 3, accurate separation and identification of multi-order frequency components: bandpass filtering is performed on the original vibration signal of each ROI to retain the expected frequency range of the cable; then variational mode decomposition method is used to adaptively decompose the filtered signal into a series of modal components with center frequencies; by analyzing the spectrum of each modal component, the estimated values ​​of the first K dominant vibration frequencies are identified and extracted to form an initial frequency set.

[0054] Step 3.1: Based on the structural design parameters, set the expected frequency range of the signal. f low , f high ],set up f low =0.1Hz, f high =1.5 f max .

[0055] Step 3.2, for each filtered signal x i ( t The variational mode decomposition (VMD) algorithm is applied. VMD adaptively decomposes the signal into its constituent parts by solving the following constrained variational problem. K Individual eigenmode functions u k ( t ):

[0056] in, It is the first One modal component, It is its center frequency. It is the Dirac function. This represents convolution. By introducing a quadratic penalty term and Lagrange multipliers, an iterative solution using the alternating direction multiplier method is employed until the convergence condition is met. (Modal number) K Ensure coverage of major frequency components by observing the spectrum or automatically determining the center frequency using the center frequency method.

[0057] Step 3.3, frequency extraction.

[0058] For each modal component obtained from the decomposition Perform a Fast Fourier Transform (FFT) to obtain its spectrum. .

[0059] Identification The frequency corresponding to the highest peak value of the medium amplitude The frequency of all modalities extracted from all ROIs. The data are then compiled to form an initial frequency observation set. ,in .

[0060] Step 4, Spatial Consistency Frequency Fusion: The initial frequency sets extracted from all ROIs are summarized to form a high-dimensional frequency observation dataset. A strategy combining principal component analysis and clustering algorithms is employed. First, the distribution characteristics of the frequency data are analyzed. Then, a density-based spatial clustering algorithm with noise is used to cluster the dominant frequency estimates, eliminating outlier frequencies caused by local image noise, sudden changes in illumination, or non-structural vibrations. The centroid frequency values ​​of each effective cluster are identified as the spatially consistent true vibration frequencies {F1, F2, ..., F...} of the photovoltaic flexible support cable structure as a whole. m}

[0061] Step 4.1, Principal Component Analysis and Data Reduction.

[0062] Will The dataset is considered a high-dimensional observation dataset. Principal component analysis (PCA) is first performed to calculate the eigenvalues ​​and eigenvectors of its covariance matrix. Principal components with a contribution rate exceeding 95% are retained. The original data is then projected into a low-dimensional space to achieve data reduction and decorrelation, resulting in the final dataset. .

[0063] Step 4.2, density-based spatial clustering.

[0064] right The density-based spatial clustering (DBSCAN) algorithm with noise is applied. This algorithm defines two parameters: neighborhood radius. and minimum points .

[0065] Set according to the expected error range of the frequency measurement. Set as DBSCAN ensures that a valid frequency cluster is supported by a sufficient number of ROI observations. It divides data points into core points, boundary points, and noise points. A cluster is a densely connected set of core and boundary points, corresponding to different orders of true vibration frequencies. Points marked as noise are outliers and are removed.

[0066] Step 4.3, Determine the actual frequency.

[0067] For the first one identified by DBSCAN 1 effective cluster The weighted centroid of all its member frequency values ​​is calculated as the true frequency of that order. :

[0068] Among them, weight This can be taken as the peak amplitude of the spectrum of the modal component corresponding to that frequency or the signal-to-noise ratio. Ultimately, an ordered, spatially consistent set of multi-order true frequencies is obtained. ,in .

[0069] Step 5, Considering boundary flexibility and environmental coupling, calculate the cable force: Establish a refined cable force calculation model suitable for photovoltaic flexible supports: F n =(n / (2L e ))*sqrt(T / ρ e )*Ψ(EI,θ,K eq ), where L e To account for the effective vibration length affected by cable clamps and connectors, ρ e To account for the equivalent linear density of the appendages, Ψ is a comprehensive correction function, whose variables include the bending stiffness EI of the cable, the elastic modulus correction coefficient of the cable at the current temperature θ, and the equivalent boundary stiffness K obtained based on the finite element model of the support or on-site calibration. eq The multi-order reliable vibration frequencies {F} obtained in step S4 n Substituting the real-time temperature θ and other known parameters into the model, the weighted least squares optimization algorithm is used for inversion and solution, and finally the high-precision real-time cable force value T is output, where the weights are allocated according to the confidence level of each frequency identification.

[0070] Step 5.1, refine the cable force calculation model.

[0071] Considering the effects of bending stiffness, boundary elastic constraints, and ambient temperature, an improved chord-beam combined vibration frequency equation is established:

[0072] in, F n Indicates the first n First vibration frequency, Hz; n Indicates the vibration order; L e Indicates the effective vibration length, in meters (m). T The force to be sought is N; ρ e It represents the equivalent linear density, in kg / m, including the mass of the cable itself and the attached materials; The comprehensive correction function is expressed as follows:

[0073] First item The second term represents the bending stiffness correction term. This indicates the temperature correction term. This represents the temperature coefficient of elastic modulus of the cable, 1 / °C. Indicates real-time temperature; Indicates reference temperature; third item This represents the boundary stiffness correction factor, obtained through modal analysis fitting of the finite element model or on-site calibration, and relates to the equivalent boundary stiffness. K eq A function of order n.

[0074] Step 5.2, multi-order frequency-weighted least squares inversion.

[0075] The multi-order frequencies identified in step 4 and its corresponding estimation order Substituting these equations into the above model, we obtain an overdetermined system of equations.

[0076] The cable forces are solved using the weighted least squares method. The optimization problem is expressed as:

[0077] Among them, weight With the The confidence level of the order frequency identification is proportional to the number of clusters in S4. The density or number of interior points is determined by this.

[0078] The above nonlinear least squares problem is solved by an iterative algorithm, and finally, a high-precision real-time cable force value is output. The model uses a correction function. The system systematically considers the unique boundary conditions, attachments, and environmental coupling effects of photovoltaic flexible supports, which significantly improves the accuracy and reliability of cable force inversion.

[0079] This invention provides a non-contact monitoring method for the cable force of a photovoltaic flexible support based on video vibration. Its core lies in achieving high-precision, non-contact measurement of cable force through a complete technology chain encompassing visual perception, signal purification, feature fusion, and physical computation. First, high-frame-rate, high-resolution cameras are deployed remotely to stably acquire clear video sequences of the cable under environmental wind-induced and operational micro-vibrations without the need for physical sensors or artificial targets. Then, the cable image is adaptively divided into Regions of Interest (ROIs). By calculating the temporal average of pixel intensity within each ROI, the two-dimensional image information is converted into a one-dimensional vibration time-history signal. Differential and wavelet denoising are used to suppress illumination drift and image noise, achieving a robust conversion from spatial motion to temporal signal. Next, the signal from each ROI is bandpass filtered, and variational mode decomposition (VMD) is used to adaptively separate multiple intrinsic modes, initially extracting estimates of the multi-order dominant frequencies reflected at each location to form an initial frequency set. To eliminate outliers caused by local noise and non-structural vibrations, a strategy combining principal component analysis and density-based spatial clustering (DBSCAN) is employed. Frequency observations from all ROIs are fused and analyzed to identify spatially consistent and dense true frequency clusters. The weighted centroid frequency of these clusters is then used as the reliable multi-order vibration frequencies of the cable structure as a whole. Finally, these multi-order frequencies are substituted into a refined mechanical model established for the photovoltaic flexible support. This model incorporates the effects of bending stiffness, equivalent boundary elastic constraints, and temperature on material properties through a comprehensive correction function system. A weighted least squares method with frequency identification confidence as the weight is used for inversion and solution, ultimately outputting high-precision real-time cable force values. This method requires no contact with the cable under test throughout the entire process, overcoming the bottlenecks of traditional sensors in terms of installation, maintenance, and cost. Through the deep integration of visual measurement with advanced signal processing, data fusion, and physical modeling technologies, it provides an innovative and reliable solution for long-term online safety monitoring of flexible support cable forces in complex environments.

[0080] Example 2 This invention proposes a non-contact monitoring system for the cable force of a photovoltaic flexible support based on video vibration, such as... Figure 2 As shown, it includes: The vibration signal extraction module is used to acquire a preprocessed video frame sequence, project the preprocessed video frame sequence along the cable body into the image, acquire multiple dynamic regions of interest, process each dynamic region of interest to generate a pixel intensity-time history signal, which serves as the original vibration signal characterizing the local lateral vibration of the cable segment. The frequency identification and extraction module is used to filter and perform variational mode decomposition on the original vibration signal, obtain modal components, analyze the spectrum of each modal component, identify and extract the estimated values ​​of the first K dominant vibration frequencies, and form an initial frequency set. The vibration frequency fusion module is used to summarize the initial frequency set to form a high-dimensional frequency observation dataset, analyze the high-dimensional frequency observation dataset, and obtain the overall, spatially consistent true vibration frequencies of each order of the photovoltaic flexible support cable structure. The support cable force acquisition module is used to establish a refined cable force calculation model, substitute the actual vibration frequencies of each order and their corresponding estimated orders into the model for inversion and solution, and output the real-time cable force value to realize non-contact monitoring of the cable force of photovoltaic flexible support.

[0081] Example 3 Please see Figure 3 As shown, the present invention also provides an electronic device 100 for a non-contact monitoring method of photovoltaic flexible support cable force based on video vibration; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0082] The memory 101 can be used to store the computer program 103. The processor 102 implements the steps of the non-contact monitoring method for the cable force of a photovoltaic flexible support based on video vibration as described in Embodiment 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device 100 (such as audio data), etc. In addition, the memory 101 may include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other non-volatile solid-state storage device.

[0083] The at least one processor 102 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or any conventional processor. The processor 102 is the control center of the electronic device 100, connecting various parts of the electronic device 100 via various interfaces and lines.

[0084] The memory 101 in the electronic device 100 stores multiple instructions to implement a non-contact monitoring method for the cable force of a photovoltaic flexible support based on video vibration, and the processor 102 can execute the multiple instructions to achieve the following: A preprocessed video frame sequence is obtained, and the preprocessed video frame sequence is projected along the cable body into the image to obtain multiple dynamic regions of interest. Each dynamic region of interest is processed to generate a pixel intensity-time history signal, which serves as the original vibration signal characterizing the local lateral vibration of the cable segment. The original vibration signal is filtered and subjected to variational mode decomposition to obtain modal components. The spectrum of each modal component is analyzed, and the estimated values ​​of the first K dominant vibration frequencies are identified and extracted to form an initial frequency set. The initial frequency set is summarized to form a high-dimensional frequency observation dataset. The high-dimensional frequency observation dataset is analyzed to obtain the overall, spatially consistent true vibration frequencies of each order of the photovoltaic flexible support cable structure. A refined cable force calculation model is established, and the actual vibration frequencies of each order and their corresponding estimated orders are substituted into the model for inversion and solution, outputting real-time cable force values, thereby realizing non-contact monitoring of the cable force of photovoltaic flexible supports.

[0085] Example 4 If the modules / units integrated in the electronic device 100 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0086] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0087] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0088] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0089] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A non-contact monitoring method for the cable force of a photovoltaic flexible support based on video vibration, characterized in that, Includes the following steps: A preprocessed video frame sequence is obtained, and the preprocessed video frame sequence is projected along the cable body into the image to obtain multiple dynamic regions of interest. Each dynamic region of interest is processed to generate a pixel intensity-time history signal, which serves as the original vibration signal characterizing the local lateral vibration of the cable segment. The original vibration signal is filtered and subjected to variational mode decomposition to obtain modal components. The spectrum of each modal component is analyzed, and the estimated values ​​of the first K dominant vibration frequencies are identified and extracted to form an initial frequency set. The initial frequency set is summarized to form a high-dimensional frequency observation dataset. The high-dimensional frequency observation dataset is analyzed to obtain the overall, spatially consistent true vibration frequencies of each order of the photovoltaic flexible support cable structure. A refined cable force calculation model is established, and the actual vibration frequencies of each order and their corresponding estimated orders are substituted into the model for inversion and solution, outputting real-time cable force values, thereby realizing non-contact monitoring of the cable force of photovoltaic flexible supports.

2. The non-contact monitoring method for photovoltaic flexible support cable force based on video vibration according to claim 1, characterized in that, The process of projecting the preprocessed video frame sequence along the cable body into the image to obtain multiple dynamic regions of interest is as follows: Along the centerline of the cable body, the cable body image region is divided into N consecutive dynamic regions of interest, each with a width of... Adaptive adjustment: in, LocalWidth (i) For the cable body in the first i The image pixel width at each location α This is a scaling factor; for curved cable segments, the ROI is divided along the tangent to the cable's centerline to ensure that the axis of the dynamically interested region is perpendicular to the direction of the cable's local vibration.

3. The non-contact monitoring method for photovoltaic flexible support cable force based on video vibration according to claim 1, characterized in that, The process of processing each dynamic region of interest to generate a pixel intensity-time history signal, which serves as the original vibration signal characterizing the local lateral vibration of the cable segment, specifically involves: For the j The first frame of the image i For each dynamic region of interest, calculate the average intensity of all pixels within that region in the green channel. I i,j : in, M This represents the total number of pixels within the ROI. G m ( i , j ) represents the intensity value of the m-th pixel in the G channel within the dynamically defined region of interest; For all video frames j =1,2,…, J Repeat this process to obtain the first... i Pixel intensity-time history signal sequence of a dynamic region of interest , For the first i The average pixel intensity of each ROI in the first frame image. For the first i The average pixel intensity of each ROI in the last frame of the image; For each pixel intensity-time history signal sequence First-order difference and wavelet threshold denoising are performed to obtain pixel intensity-time history signals.

4. The non-contact monitoring method for photovoltaic flexible support cable force based on video vibration according to claim 1, characterized in that, The process of filtering and performing variational mode decomposition on the original vibration signal to obtain modal components, analyzing the spectrum of each modal component, identifying and extracting the estimated values ​​of the first K dominant vibration frequencies, and forming an initial frequency set, specifically involves: Bandpass filtering is performed on the original vibration signal of each dynamic region of interest to obtain the expected frequency range of the signal. f low , f high ];in, f low To be the minimum expected frequency, f high This represents the maximum expected frequency. For each filtered signal x i ( t The variational mode decomposition algorithm is applied to adaptively decompose the signal into... K eigenmode functions u k ( t ): in, It is the first One modal component, It is its center frequency. It is the Dirac function. The convolution is represented by a quadratic penalty term and Lagrange multipliers, and the solution is iteratively obtained using the alternating direction multiplier method until the convergence condition is met; the number of modes... K Determined automatically by observing the spectrum or using the center frequency method; For each modal component obtained from the decomposition Perform a Fast Fourier Transform to obtain its spectrum. Identify the spectrum The frequency corresponding to the highest peak value of the medium amplitude The frequency of all modes extracted from all dynamic regions of interest. The data are then compiled to form an initial frequency observation set. ,in ; For the first frequency observation, This is the second frequency observation. For the first Each frequency observation.

5. The non-contact monitoring method for photovoltaic flexible support cable force based on video vibration according to claim 1, characterized in that, The aggregated initial frequency set constitutes a high-dimensional frequency observation dataset. Analysis of this high-dimensional frequency observation dataset yields the spatially consistent true vibration frequencies of each order for the entire photovoltaic flexible support cable structure. Specifically: Initial frequency observation set Treating the dataset as a high-dimensional observation dataset, principal component analysis is performed to calculate the eigenvalues ​​and eigenvectors of the dataset's covariance matrix. The original data is then projected onto the space corresponding to the principal components to obtain the dataset. ; For dataset A density-based spatial clustering algorithm is applied to divide data points into core points, boundary points, and noise points. A set of density-connected core points and boundary points constitutes a cluster, corresponding to different orders of true vibration frequencies. Points marked as noise are outliers and are removed. The spatial clustering algorithm identifies the first 1 effective cluster Calculate the weighted centroid of all member frequency values ​​as the true frequency of that order. : Among them, weight This involves determining the peak amplitude or signal-to-noise ratio of the modal component corresponding to that frequency, ultimately resulting in an ordered, spatially consistent set of multi-order true frequencies. ,in, , The first vibration frequency of the cable is... The second-order vibration frequency, It is the Mth order vibration frequency.

6. The non-contact monitoring method for the cable force of photovoltaic flexible support based on video vibration according to claim 1, characterized in that, The establishment of a refined cable force calculation model specifically involves: in, F n Indicates the first n First vibration frequency; n Indicates the vibration order; L e Indicates the effective vibration length; T Indicates the force to be claimed; ρ e It represents the equivalent linear density, including the mass of the cable itself and the attached material; The comprehensive correction function is expressed as follows: This indicates a correction term for bending stiffness; This indicates the temperature correction term. This indicates the temperature coefficient of the elastic modulus of the cable. Indicates real-time temperature; Indicates reference temperature; This represents the boundary stiffness correction factor, obtained through modal analysis fitting of the finite element model or on-site calibration, and relates to the equivalent boundary stiffness. K eq A function of order n.

7. The non-contact monitoring method for the cable force of photovoltaic flexible support based on video vibration according to claim 6, characterized in that, The process involves substituting the actual vibration frequencies and their corresponding estimated orders into the model for inversion and solving, outputting real-time cable force values. Specifically: The true vibration frequencies of each order and its corresponding estimation order Substitute the values ​​into the refined cable force calculation model used for photovoltaic flexible supports; The cable forces are solved using the weighted least squares method. The optimization problem is expressed as: Among them, weight With the The confidence level of order frequency identification is proportional to that of clusters. The density or number of interior points determines the cable force; by iteratively solving the above nonlinear least squares problem, the real-time cable force value is output. .

8. A non-contact monitoring system for the cable force of a photovoltaic flexible support based on video vibration, characterized in that, include: The vibration signal extraction module is used to acquire a preprocessed video frame sequence, project the preprocessed video frame sequence along the cable body into the image, acquire multiple dynamic regions of interest, process each dynamic region of interest to generate a pixel intensity-time history signal, which serves as the original vibration signal characterizing the local lateral vibration of the cable segment. The frequency identification and extraction module is used to filter and perform variational mode decomposition on the original vibration signal, obtain modal components, analyze the spectrum of each modal component, identify and extract the estimated values ​​of the first K dominant vibration frequencies, and form an initial frequency set. The vibration frequency fusion module is used to summarize the initial frequency set to form a high-dimensional frequency observation dataset, analyze the high-dimensional frequency observation dataset, and obtain the overall, spatially consistent true vibration frequencies of each order of the photovoltaic flexible support cable structure. The support cable force acquisition module is used to establish a refined cable force calculation model, substitute the actual vibration frequencies of each order and their corresponding estimated orders into the model for inversion and solution, and output the real-time cable force value to realize non-contact monitoring of the cable force of photovoltaic flexible support.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the non-contact monitoring method for the cable force of photovoltaic flexible support based on video vibration as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the non-contact monitoring method for the cable force of photovoltaic flexible support based on video vibration as described in any one of claims 1 to 7.

Images