Method, device and electronic equipment for detecting abnormalities in wind turbine blades

Abstract

This invention discloses a method, device, and electronic equipment for detecting anomalies in wind turbine blades. The method includes: acquiring initial vibration data corresponding to multiple laser measuring points on the target blade surface, position coordinates and attitude data of a UAV at multiple sampling time periods, and pointing data of a gimbal at multiple sampling time periods; determining the target vibration data corresponding to each of the multiple laser measuring points; performing feature extraction processing on the target vibration data corresponding to each of the multiple laser measuring points to obtain vibration feature data corresponding to each of the multiple laser measuring points; determining whether there are abnormal laser measuring points among the multiple laser measuring points; and, if abnormal laser measuring points exist, determining the anomaly distribution map of the target blade based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measuring points. This invention solves the technical problem of low positioning accuracy of wind turbine blade anomalies caused by UAV motion interference.

Description

Technical Field

[0001] This invention relates to the fields of new energy and energy conservation technology, and more specifically, to a method, device, and electronic equipment for detecting abnormalities in wind turbine blades. Background Technology

[0002] As the most complex and structurally vulnerable key component of a wind turbine generator set, wind turbine blades are exposed to a complex and ever-changing natural environment for extended periods, making them highly susceptible to internal damage such as cracks, delamination, and debonding due to fatigue, lightning strikes, ice erosion, rain erosion, and material aging. Such damage often presents no surface signs in its early stages, but it leads to a reduction in local stiffness, significantly altering the blade's natural frequency, mode shape, and damping ratio. Therefore, capturing subtle changes in blade vibration characteristics through high-precision, non-contact vibration monitoring technology to achieve intelligent early warning of damage has become a research direction in the field of intelligent wind power operation and maintenance. The significant shortcomings of related technologies in wind turbine blade anomaly detection methods are mainly reflected in the following aspects:

[0003] Vibration monitoring of blades using sensors such as adhesive accelerometers, fiber optic gratings, or strain gauges can obtain local vibration data with high signal-to-noise ratios, but these methods require manual installation at height while the blade is shut down, resulting in complex wiring, high costs, and the acquisition of vibration signals from only a limited number of discrete measurement points, failing to reflect the vibration distribution across the entire blade field. Ground-based laser Doppler vibration meters, while possessing micron-level displacement resolution, require installation tens of meters away from the blades. Limited by terrain, vegetation, obstacles, and wind field layout, it is difficult to achieve comprehensive coverage of all blades. During long-distance propagation, the laser beam is susceptible to atmospheric turbulence, temperature and humidity gradients, and scattering by dust and water vapor, leading to signal distortion, a sharp drop in signal-to-noise ratio, and poor measurement stability. Using drones for visual photography inspections can only detect visible defects on the blade surface, failing to perceive stiffness changes caused by internal damage, and is greatly affected by light and dirt. Although wind turbine blade anomaly detection methods in related technologies can provide the acquisition of local vibration signals and visual identification of surface defects within a certain range, the failure to effectively separate the drone's own motion noise from the actual vibration of the blades leads to low accuracy in locating wind turbine blade anomalies.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This invention provides a method, device, and electronic equipment for detecting anomalies in wind turbine blades, in order to at least solve the technical problem of low positioning accuracy of wind turbine blade anomalies caused by interference from drone movement.

[0006] According to one aspect of the present invention, an anomaly detection method for wind turbine blades is provided, comprising: acquiring initial vibration data corresponding to multiple laser measuring points on the surface of a target blade in a wind turbine, collected at multiple sampling time periods; position coordinates and attitude data of a drone at each of the multiple sampling time periods; and pointing data of a gimbal at each of the multiple sampling time periods. The initial vibration data is obtained by sampling corresponding laser measuring points along a preset inspection path using a laser vibration meter in the drone. Multiple sampling time periods correspond one-to-one with multiple laser measuring points. The position coordinates represent the position coordinates of the drone relative to the center of the wind turbine rotor. The gimbal is a rotating platform installed in the drone for controlling the laser beam sampling direction. The initial vibration data, along with the position coordinates, attitude data, and pointing data corresponding to each of the multiple sampling periods, are used to obtain target vibration data for each of the multiple laser measuring points. The target vibration data represents the vibration data caused by the vibration of the target blade after excluding interference from UAV movement. Feature extraction processing is performed on the target vibration data corresponding to each of the multiple laser measuring points to obtain vibration feature data for each laser measuring point. Based on the vibration feature data corresponding to each of the multiple laser measuring points and preset health feature data, it is determined whether there are any abnormal laser measuring points among the multiple laser measuring points. If abnormal laser measuring points are found, an abnormal distribution map of the target blade is determined based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measuring points.

[0007] According to another aspect of the present invention, an anomaly detection device for wind turbine blades is also provided, comprising: a data acquisition module, configured to acquire initial vibration data corresponding to multiple laser measuring points on the surface of a target blade in a wind turbine, collected during multiple sampling periods; position coordinates and attitude data of a drone during each of the multiple sampling periods; and pointing data of a gimbal during each of the multiple sampling periods. The initial vibration data is obtained by sampling corresponding laser measuring points along a preset inspection path using a laser vibration meter in the drone. Multiple sampling periods correspond one-to-one with multiple laser measuring points. The position coordinates represent the position coordinates of the drone relative to the center of the wind turbine rotor. The gimbal is a rotating platform installed in the drone for controlling the laser beam sampling direction. A target vibration data determination module is configured to determine the target vibration data based on the initial vibration data corresponding to each of the multiple laser measuring points. The system includes multiple sampling time periods, corresponding position coordinates, attitude data, and pointing data to obtain target vibration data for each laser measuring point. The target vibration data represents the vibration data caused by the vibration of the target blade after eliminating UAV motion interference. A vibration feature data determination module is used to perform feature extraction processing on the target vibration data corresponding to each laser measuring point to obtain vibration feature data for each laser measuring point. An abnormal laser measuring point judgment module is used to determine whether there are abnormal laser measuring points among the multiple laser measuring points based on the vibration feature data corresponding to each laser measuring point and preset health feature data. An abnormal distribution map determination module is used to determine the abnormal distribution map of the target blade based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measuring points when abnormal laser measuring points exist among the multiple laser measuring points.

[0008] According to another aspect of the present invention, a non-volatile storage medium is also provided, the non-volatile storage medium storing a plurality of instructions adapted for loading by a processor and executing any one of the methods for detecting anomalies in wind turbine blades.

[0009] According to another aspect of the present invention, an electronic device is also provided, including one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the wind turbine blade anomaly detection method described in any one of the present invention.

[0010] According to another aspect of the present invention, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps of the method for anomaly detection of wind turbine blades as described in any one of the present invention.

[0011] In this embodiment of the invention, initial vibration data corresponding to multiple laser measuring points on the surface of the target blade of a wind turbine generator set are acquired at multiple sampling time periods, along with the position coordinates and attitude data of the UAV at each of the multiple sampling time periods, and the pointing data of the gimbal at each of the multiple sampling time periods. The initial vibration data is obtained by sampling the corresponding laser measuring points along a preset inspection path using a laser vibration meter in the UAV. Multiple sampling time periods correspond one-to-one with multiple laser measuring points. The position coordinates represent the position coordinates of the UAV relative to the center of the wind turbine rotor. The gimbal is a rotating platform installed in the UAV to control the laser beam sampling direction. Based on the initial vibration data corresponding to each of the multiple laser measuring points, and the position coordinates, attitude data, and pointing data corresponding to each of the multiple sampling time periods, target vibration data corresponding to each of the multiple laser measuring points is obtained. The target vibration data represents the vibration data caused by the vibration of the target blade after excluding interference from the UAV's motion. Feature extraction processing is performed on the target vibration data corresponding to each of the multiple laser measuring points to obtain multiple laser beams. The system obtains vibration characteristic data corresponding to each optical measuring point; based on the vibration characteristic data corresponding to each of the multiple laser measuring points, as well as preset health characteristic data, it determines whether there are abnormal laser measuring points among the multiple laser measuring points; in the case of abnormal laser measuring points, based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measuring points, it determines the abnormal distribution map of the target blade. This achieves the goal of accurately obtaining the abnormal distribution map of the target blade by acquiring the initial vibration data of multiple laser measuring points, the position coordinates, attitude data, and pointing data of multiple sampling periods, and extracting features from the target vibration data to determine whether there are abnormal laser measuring points. When abnormal laser measuring points are found, based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measuring points, it determines the position of the abnormal measuring points on the target blade, thereby achieving the technical effect of improving the abnormal positioning accuracy of wind turbine blades and solving the technical problem of low abnormal positioning accuracy of wind turbine blades caused by UAV motion interference. Attached Figure Description

[0012] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0013] Figure 1 This is a flowchart of an anomaly detection method for wind turbine blades according to an embodiment of the present invention;

[0014] Figure 2 This is a flowchart of a method for determining target vibration data corresponding to multiple optional laser measuring points according to an embodiment of the present invention;

[0015] Figure 3 This is a flowchart of an optional method for detecting anomalies in wind turbine blades according to an embodiment of the present invention;

[0016] Figure 4 This is a schematic diagram of an abnormality detection device for wind turbine blades according to an embodiment of the present invention;

[0017] Figure 5 This is a schematic diagram of an electronic device for detecting abnormalities in wind turbine blades according to an embodiment of the present invention. Detailed Implementation

[0018] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0019] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] According to an embodiment of the present invention, a method for detecting anomalies in wind turbine blades is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0021] Figure 1 This is a flowchart of an anomaly detection method for wind turbine blades according to an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes the following steps:

[0022] Step S102: Acquire the initial vibration data corresponding to multiple laser measuring points on the surface of the target blade of the wind turbine generator set collected during multiple sampling periods, the position coordinates and attitude data of the UAV corresponding to each of the multiple sampling periods, and the pointing data of the gimbal corresponding to each of the multiple sampling periods. The initial vibration data is obtained by sampling the corresponding laser measuring points along the preset inspection path using the laser vibration meter in the UAV. The multiple sampling periods correspond one-to-one with the multiple laser measuring points. The position coordinates represent the position coordinates of the UAV relative to the center of the wind turbine generator set. The gimbal is a rotating platform installed in the UAV to control the sampling direction of the laser beam.

[0023] Optionally, by equipping a UAV with a laser vibrometer and a quantum inertial navigation system, and integrating a high-precision two-dimensional gimbal, a non-contact dynamic vibration monitoring system for target blades can be constructed. During the inspection process, instead of performing instantaneous measurements at a single moment or fixed point, continuous and synchronous data acquisition is conducted along a pre-set inspection path (such as multiple hovering points or continuous scanning areas along the blade span) at multiple sampling periods, targeting multiple laser measurement points on the blade surface. Each sampling period corresponds to a specific laser measurement point location, ensuring that vibration data spatially covers the entire blade surface and forms a continuous dynamic sequence in time. During each sampling period, the laser vibrometer collects the initial vibration data of that laser measurement point in real time. This initial vibration data is the minute displacement changes on the blade surface reflected by the laser beam echo, containing a mixed signal of the target blade's actual vibration response and the UAV's own motion. Simultaneously, the quantum inertial measurement unit installed on the drone continuously outputs the drone's position coordinates and attitude data relative to the wind turbine's rotor center at the sampling moment. The attitude data includes the drone's yaw, pitch, and roll angles during the corresponding sampling period, thus accurately characterizing the drone's motion in space. Meanwhile, a high-precision encoder installed in the drone's gimbal synchronously records the gimbal's pointing data relative to the drone's body. This pointing data includes the gimbal's pitch and yaw angles pointing towards the corresponding laser measurement point during the corresponding sampling period. These multi-source heterogeneous data can be synchronously acquired via an onboard computer, forming a four-dimensional correlated dataset of "measurement point—time—attitude—pointing." This overcomes the problems of sparse measurement points, limited ground-based laser vibration measurement distance, and the inability of visual inspection to detect structural damage inherent in contact-based measurements.

[0024] Step S104: Based on the initial vibration data corresponding to each of the multiple laser measuring points, and the position coordinates, attitude data and pointing data corresponding to each of the multiple sampling time periods, the target vibration data corresponding to each of the multiple laser measuring points is obtained. The target vibration data represents the vibration data caused by the vibration of the target blade after excluding the interference of the UAV motion.

[0025] Optionally, based on the initial vibration data corresponding to multiple laser measurement points, as well as the UAV's position coordinates, attitude data, and gimbal pointing data during the corresponding sampling period, a precise physical separation model can be constructed to achieve high-precision physical stripping of UAV motion interference, thereby extracting the pure vibration signal caused only by the elastic vibration of the target blade, i.e., the target vibration data. This process can achieve real-time, adaptive, and drift-free compensation for UAV motion noise, solving the technical challenges of signal distortion and feature submersion caused by UAV platform jitter in UAV laser vibration measurement in related technologies.

[0026] In one alternative embodiment, Figure 2 This is a flowchart of a method for determining target vibration data corresponding to multiple optional laser measuring points according to an embodiment of the present invention, as shown below. Figure 2 As shown, based on the initial vibration data corresponding to multiple laser measuring points, and the position coordinates, attitude data, and pointing data corresponding to multiple sampling periods, the target vibration data corresponding to each of the multiple laser measuring points is obtained, including:

[0027] Step S202: Based on the attitude data and pointing data corresponding to each of the multiple sampling time periods, obtain the laser beam pointing vector corresponding to each of the multiple laser measurement points. The laser beam pointing vector is used to indicate the direction in which the laser beam points to the corresponding laser measurement point during the corresponding sampling time period.

[0028] Step S204: Based on the position coordinates corresponding to each of the multiple sampling time periods and the laser beam pointing vectors corresponding to each of the multiple laser measuring points, the displacement projection data corresponding to each of the multiple laser measuring points is obtained. The displacement projection data represents the interference vibration data obtained by the UAV movement at the corresponding laser measuring point.

[0029] Step S206: Based on the initial vibration data and displacement projection data corresponding to each of the multiple laser measuring points, obtain the target vibration data corresponding to each of the multiple laser measuring points.

[0030] Optionally, firstly, using the UAV's attitude data and gimbal pointing data for each sampling period, a coordinate transformation can be performed to obtain the real-time pointing vector of the laser beam in inertial space, i.e., the laser beam pointing vector. This vector precisely describes the direction of the laser beam pointing to the corresponding laser measurement point in three-dimensional space. It dynamically evolves with changes in UAV attitude and gimbal rotation, ensuring that the laser is always accurately aligned with the preset laser measurement point on the target blade surface, maintaining optical closed-loop tracking even under wind disturbances and platform sway. Then, combining the UAV's position coordinates at the corresponding laser measurement point, which are output in real-time by the quantum inertial navigation system, these coordinates are projected along the corresponding laser beam pointing vector to calculate the displacement component of the UAV's overall motion in the laser measurement direction, i.e., the displacement projection data. The displacement projection data for any laser measurement point can be calculated as follows: ,in, This represents the displacement projection data of any laser measurement point. This represents the laser beam pointing vector for any sampling period. The vector representation of the position coordinates for any sampling period, where t represents the index of any sampling period, shows that the calculated displacement projection data is essentially a false vibration interference signal caused by the translation and rotation of the UAV, rather than the elastic response of the target blade itself. Finally, by subtracting the corresponding displacement projection data from the original vibration signal (containing interference) collected by the laser vibrometer, the pure signal caused solely by the vibration of the target blade structure—the target vibration data—is obtained. This embodiment's method achieves high-fidelity, real-time, and drift-free elimination of UAV motion noise, overcoming the defects of vibration signal distortion and feature confusion caused by UAV motion.

[0031] Step S106: Perform feature extraction processing on the target vibration data corresponding to each of the multiple laser measuring points to obtain the vibration feature data corresponding to each of the multiple laser measuring points.

[0032] Optionally, feature extraction processing is performed on the target vibration data corresponding to each laser measuring point, i.e., the actual vibration signal of the target blade after removing interference from UAV motion, to obtain vibration feature data for each laser measuring point. The obtained vibration feature data can directly reflect the mass distribution of the local structure of the target blade and is a sensitive indicator for identifying anomalies (such as cracks, delamination, debonding, and other early damage). By independently extracting vibration feature data for each laser measuring point, high-density, spatially resolved anomaly scanning of the target blade surface can be achieved, unlike related technologies that rely on only a few points or coarse-grained assessments of the overall response. In single-point scanning mode, this method can generate a continuously distributed feature profile along the blade span; in array parallel mode, it can construct a vibration feature distribution map covering the entire surface of the target blade. This refined feature acquisition method can not only determine whether there are anomalies in the target blade but also accurately locate the anomalies, thereby significantly improving the sensitivity and diagnostic accuracy of early damage identification for the target blade.

[0033] In an optional embodiment, when the vibration feature data includes the natural frequency and damping ratio of the corresponding laser measuring point, feature extraction processing is performed on the target vibration data corresponding to each of the multiple laser measuring points to obtain vibration feature data corresponding to each of the multiple laser measuring points. This includes: denoising the target vibration data corresponding to each of the multiple laser measuring points to obtain denoised vibration data corresponding to each of the multiple laser measuring points; and performing time-frequency domain transformation processing on the denoised vibration data corresponding to each of the multiple laser measuring points to obtain time-frequency energy distribution maps corresponding to each of the multiple laser measuring points. The time-frequency domain transformation is a short-time Fourier transform or a wavelet transform, and the time-frequency energy distribution map is used to indicate the vibration energy distribution characteristics of the corresponding laser measuring point in the time-frequency domain. The time-frequency energy distribution maps corresponding to multiple laser measuring points are obtained to determine the natural frequencies of each laser measuring point. The natural frequencies represent the resonant frequencies of the target blade at the corresponding laser measuring point. Hilbert transforms are applied to the denoised vibration data corresponding to each laser measuring point to obtain amplitude envelope maps. The horizontal axis of the amplitude envelope map represents multiple sampling times within the sampling period of the corresponding laser measuring point, and the vertical axis represents the vibration displacement amplitude at each sampling time. Based on the amplitude envelope maps corresponding to each laser measuring point, the damping ratios are obtained. The damping ratios are used to quantify the dissipation rate of vibration energy of the target blade at the corresponding laser measuring point.

[0034] Optionally, when the vibration characteristic data includes natural frequencies and damping ratios, a feature extraction process can be implemented on the target vibration data at each laser measuring point to achieve high-precision quantification of the local structural dynamic parameters of the target blade. Specifically, firstly, adaptive denoising processing is performed on the target vibration data at each laser measuring point. Methods such as frequency domain filtering or wavelet thresholding are used to eliminate residual environmental noise interference, ensuring the signal purity for subsequent analysis and obtaining denoised vibration data. Subsequently, time-frequency domain transformation is performed on the denoised vibration data, preferably short-time Fourier transform or wavelet transform, to generate a time-frequency energy distribution map corresponding to each laser measuring point. This map uses multiple sampling times within the corresponding sampling period as the horizontal axis, the frequencies corresponding to each sampling time as the vertical axis, and the vibration energy amplitude as the color scale, intuitively revealing the energy accumulation characteristics of different frequency components of the target blade over time during operation. By identifying this time-frequency energy distribution map, the natural frequency of the target blade at that laser measuring point can be accurately located. This natural frequency directly reflects the distribution of local stiffness and mass; even a small deviation corresponds to material degradation or structural damage. Furthermore, by performing a Hilbert transform on the denoised vibration signal, the amplitude envelope map of the corresponding laser measurement point is extracted. This amplitude envelope map uses multiple sampling times included in the corresponding sampling period as the horizontal axis and the vibration displacement amplitude envelope corresponding to each sampling time as the vertical axis, thus fully depicting the temporal trend of vibration attenuation. Based on this amplitude envelope map, the damping ratio of the laser measurement point can be accurately retrieved. The damping ratio is used to quantitatively characterize the energy dissipation capacity caused by internal friction, microcrack friction, or delamination interface slippage of the material. An abnormal increase in the damping ratio is a typical indicator of early structural damage. Since each laser measurement point independently completes the method of this embodiment, a high-density "frequency-damping" spectrum covering the entire surface of the target blade can be constructed, forming a perception capability of the target blade's health status.

[0035] In one optional embodiment, the natural frequencies corresponding to each of the multiple laser measuring points are obtained based on the time-frequency energy distribution maps corresponding to each of the multiple laser measuring points, including: taking the vibration energy amplitude greater than a preset first amplitude threshold in the time-frequency energy distribution map of any laser measuring point among the multiple laser measuring points as the target vibration energy amplitude of any laser measuring point; determining the frequency corresponding to the target vibration energy amplitude as the natural frequency of any laser measuring point; and obtaining the natural frequencies corresponding to each of the multiple laser measuring points by using the method of obtaining the natural frequency of any laser measuring point.

[0036] Optionally, to accurately extract the natural frequency from the time-frequency energy distribution map of each laser measuring point, an adaptive peak detection mechanism based on energy threshold screening is adopted. Specifically, for the time-frequency energy distribution map of any laser measuring point, a preset first amplitude threshold is first set. This preset first amplitude threshold can be calibrated according to the background noise level and is used to filter out low-energy random noise disturbances and interference components of non-structural resonance. Subsequently, only vibration energy amplitudes in the time-frequency energy distribution map that exceed the preset first amplitude threshold are retained and defined as target vibration energy amplitudes. Further, the frequency coordinates corresponding to the target vibration energy amplitude are located, and this frequency is determined as the natural frequency of the measuring point. By repeatedly applying the method of this embodiment, the corresponding natural frequency of each laser measuring point is extracted independently and in parallel, forming a high-density natural frequency distribution map covering the entire surface of the blade. In related technologies, subharmonics, harmonics, or external vibration sources are often mistakenly identified as natural frequencies due to noise interference, leading to misjudgment. However, this method, through energy threshold screening, can effectively shield low-energy spurious peaks and ensure that the extracted frequencies originate only from the structure-dominated resonance response.

[0037] In one optional embodiment, the damping ratio corresponding to each of the multiple laser measuring points is obtained based on the amplitude envelope diagrams corresponding to each of the multiple laser measuring points, including: taking the vibration displacement amplitude greater than a preset second amplitude threshold in the amplitude envelope diagram of any laser measuring point among the multiple laser measuring points as the target vibration displacement amplitude of any laser measuring point; obtaining the damping ratio of any laser measuring point based on the target vibration displacement amplitude and the vibration displacement amplitude at a predetermined sampling time after the target vibration displacement amplitude; and obtaining the damping ratio corresponding to each of the multiple laser measuring points by using the method of obtaining the damping ratio of any laser measuring point.

[0038] Optionally, to accurately calculate the damping ratio from the amplitude envelope of each laser measuring point, an adaptive algorithm based on dynamic energy attenuation analysis can be used. First, for the amplitude envelope of any laser measuring point, a preset second amplitude threshold is set. This preset second amplitude threshold is used to remove low-amplitude noise trails, measurement drift, or unsteady disturbance segments from the amplitude envelope, retaining only the vibration displacement amplitude dominated by the free decaying vibration of the target blade structure and exhibiting a clear attenuation trend, defined as the target vibration displacement amplitude. Then, after this target vibration displacement amplitude, vibration displacement amplitudes at predetermined sampling times are selected, and the logarithmic decay rate between the two vibration displacement amplitudes is calculated. Combined with the sampling time interval, the damping ratio is calculated according to the damping ratio calculation formula (e.g., logarithmic decay method). =(1 / n)×ln(X1 / X n ),in, X represents the damping ratio at any laser measuring point, X1 represents the target vibration displacement amplitude at any laser measuring point, X nThe amplitude of vibration displacement is the amplitude of vibration at a predetermined sampling time after the target vibration displacement amplitude. n represents the sampling interval between the target vibration displacement amplitude and the vibration displacement amplitude at the predetermined sampling time after the target vibration displacement amplitude. The damping ratio of any laser measuring point is accurately calculated. By applying the method of this embodiment to all laser measuring points, the damping ratio distribution of each laser measuring point on the entire surface of the target blade can be obtained simultaneously, thereby achieving high sensitivity and spatially resolved quantification of the local energy dissipation characteristics of the target blade. As an indicator of microscopic damage within the material, the damping ratio is extremely sensitive to non-stiffness damage such as crack friction, fiber-matrix debonding, and delamination slip.

[0039] Step S108: Based on the vibration characteristic data corresponding to each of the multiple laser measuring points and the preset health characteristic data, determine whether there are any abnormal laser measuring points among the multiple laser measuring points.

[0040] Optionally, the vibration feature data extracted from each laser measuring point is compared with preset health feature data to intelligently determine whether there are abnormal laser measuring points. The preset health feature data can be derived from historical inspection records of the target blade in a new or known health state, design simulation data, or a baseline database (health baseline database) constructed based on statistics from multiple units of the same model. By independently evaluating the vibration feature data of each laser measuring point, if the vibration feature data of a certain laser measuring point exceeds the range of the preset health feature data, the laser measuring point is marked as an abnormal laser measuring point. This judgment mechanism can achieve a leap from single-point anomaly detection to intelligent diagnosis of structural health status. Related technologies rely only on a single frequency offset, which is easily affected by environmental factors such as wind speed fluctuations and temperature changes, resulting in a high false alarm rate. In contrast, the method in this embodiment, through a composite diagnostic logic of multi-feature joint criteria, spatial distribution modeling, and dynamic baseline comparison, can significantly improve the accuracy and anti-interference capability of the judgment.

[0041] In an optional embodiment, when the vibration characteristic data includes natural frequency and damping ratio, and the preset health characteristic data includes preset natural frequency and preset damping ratio, based on the vibration characteristic data corresponding to each of the multiple laser measuring points and the preset health characteristic data, it is determined whether there are any abnormal laser measuring points among the multiple laser measuring points. This includes: obtaining the natural frequency deviation corresponding to each of the multiple laser measuring points based on the natural frequency and preset natural frequency corresponding to each of the multiple laser measuring points; obtaining the damping ratio deviation corresponding to each of the multiple laser measuring points based on the damping ratio and preset damping ratio corresponding to each of the multiple laser measuring points; determining any laser measuring point as an abnormal laser measuring point if the natural frequency deviation of any laser measuring point is greater than the preset frequency deviation, or if the damping ratio deviation of any laser measuring point is greater than the preset damping ratio deviation; or determining that there are no abnormal laser measuring points among the multiple laser measuring points if the natural frequency deviations corresponding to each of the multiple laser measuring points are all less than or equal to the preset frequency deviation, or if the damping ratio deviations corresponding to each of the multiple laser measuring points are all less than or equal to the preset damping ratio deviation.

[0042] Optionally, when the vibration characteristic data includes natural frequency and damping ratio, and the preset health characteristic data corresponds to preset natural frequency and preset damping ratio based on historical or design benchmarks, a precise and robust anomaly determination is made for the structural health status of each laser measuring point. Specifically, firstly, the natural frequency deviation between the corresponding natural frequency and the preset natural frequency is calculated for each laser measuring point, and simultaneously, the damping ratio deviation between the corresponding damping ratio and the preset damping ratio is calculated for each laser measuring point. Subsequently, preset thresholds are set for the natural frequency deviation and the damping ratio deviation, respectively. The preset frequency deviation is used to capture frequency drops caused by local stiffness degradation (such as cracks), and the preset damping ratio deviation is used to identify damping increases caused by enhanced energy dissipation (such as delamination or debonding). If the natural frequency deviation of any laser measuring point exceeds the preset frequency threshold, or its damping ratio deviation exceeds the preset damping threshold, the laser measuring point is determined to be an abnormal laser measuring point, thus fully taking into account the early signs of different types of damage. Only when the natural frequency deviation of all laser measurement points is within the allowable range and all damping ratio deviations are within the limits can the entire target blade be judged to be without abnormality, thus avoiding misjudgment due to fluctuations in a single parameter.

[0043] Step S110: In the case of abnormal laser measuring points among multiple laser measuring points, an abnormal distribution map of the target blade is determined based on the position coordinates, attitude data and pointing data corresponding to the abnormal laser measuring points.

[0044] Optionally, when abnormal laser measuring points exist among multiple laser measuring points, instead of simply marking these abnormal laser measuring points in isolation, the system fully utilizes the real-time output of the UAV's position coordinates, attitude data, and gimbal pointing data from the quantum inertial navigation system. Through spatial geometric transformation, the vibration characteristic data of each abnormal laser measuring point is precisely bound to its absolute coordinates in three-dimensional space, ultimately obtaining an anomaly distribution map of the target blade. This anomaly distribution map can clearly show whether the anomaly is concentrated in a local area (such as a strip-shaped high-damping region on the mid-section suction surface) or dispersed (such as multiple frequency drops); whether it exhibits a gradient evolution trend along the spanwise or chordwise direction; and whether it spatially overlaps with weak areas of the blade structure (such as the blade root transition zone or the end of the leading edge protective layer). This visualization of spatial correlation allows maintenance personnel to intuitively judge the morphology of damage, such as single-point cracks, linear delamination, or large-scale material degradation, providing a direct basis for formulating maintenance strategies.

[0045] In an optional embodiment, when there are multiple abnormal laser measurement points, and when an abnormal laser measurement point exists among the multiple laser measurement points, an abnormal distribution map of the target blade is determined based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measurement points. This includes: obtaining the first position coordinates corresponding to each of the multiple abnormal laser measurement points based on the target distance, position coordinates, attitude data, and pointing data corresponding to each of the multiple abnormal laser measurement points, wherein the target distance represents the straight-line distance between the corresponding abnormal laser measurement point and the UAV, and the first position coordinates represent the position coordinates of the corresponding abnormal laser measurement point relative to the center of the wind turbine; mapping the first position coordinates corresponding to each of the multiple abnormal laser measurement points onto a preset three-dimensional blade model to obtain the second position coordinates corresponding to each of the multiple abnormal laser measurement points, wherein the preset three-dimensional blade model represents a three-dimensional geometric model constructed based on the parameters of the target blade, used to indicate the mapping relationship between the coordinate system corresponding to the center of the wind turbine and the blade coordinate system corresponding to the target blade; and obtaining the abnormal distribution map based on the second position coordinates corresponding to each of the multiple abnormal laser measurement points.

[0046] Optionally, when multiple abnormal laser measurement points exist, the vibration anomalies at these points are not simply listed. Instead, a complete spatial distribution map of the damage on the target blade surface is constructed through high-precision spatial coordinate chain calculation. First, using the target distance for each abnormal measurement point—the straight-line distance between the UAV and the target blade surface—as a basis, and combining the UAV's position coordinates and attitude data output in real time by the quantum inertial navigation system, as well as the gimbal's pointing data, the first position coordinates of each abnormal laser measurement point in the global inertial coordinate system are accurately calculated using the spatial rigid body transformation formula. The first position coordinates of any abnormal laser measurement point can be obtained as follows: ,in, This represents the first position coordinates of any abnormal laser measurement point. This indicates the position coordinates corresponding to any abnormal laser measurement point. This represents the attitude data corresponding to any abnormal laser measurement point. This represents the pointing data corresponding to any abnormal laser measurement point. This represents the target distance corresponding to any abnormal laser measurement point. Subsequently, these global coordinates are mapped to a preset 3D digital model of the wind turbine blade. This preset 3D blade model is a high-precision model constructed based on the geometric parameters of the target blade (including but not limited to length, cross-sectional profile, blade root installation angle, and torsional distribution). The preset 3D blade model internally incorporates a strict geometric transformation relationship between the wind turbine center coordinate system and the blade local coordinate system (such as coordinate origin alignment and axial rotation mapping). Through the coordinate transformation engine of this preset 3D blade model, the first position coordinates of each abnormal point are accurately converted to the second position coordinates in the blade coordinate system. The origin of the blade coordinate system is the center of the target blade root; the positive direction of the horizontal axis of the blade coordinate system is the direction from the blade root to the blade tip; the positive direction of the vertical axis of the blade coordinate system is the direction from the leading edge to the trailing edge of the target blade; and the positive direction of the vertical axis of the blade coordinate system is the direction from the pressure surface to the suction surface of the target blade. The origin of the wind turbine's central coordinate system is the center of the wind turbine. The positive direction of the horizontal axis of the wind turbine's central coordinate system points to the prevailing wind direction of the wind turbine generator, the positive direction of the vertical axis of the wind turbine's central coordinate system is perpendicular to the prevailing wind direction, and the positive direction of the vertical axis of the wind turbine's central coordinate system is perpendicular to the wind turbine's rotation plane and points upwards along the wind turbine's axis. This mapping process ensures that vibration data from different inspection cycles, different UAV flight paths, and even different weather conditions can be accurately aligned within a unified blade geometry framework, achieving data comparability across time and platforms. Finally, the second position coordinates of all abnormal measuring points are superimposed onto the surface of the 3D blade model in the form of density heatmaps or contour lines to form an anomaly distribution map. This anomaly distribution map not only marks the location of abnormal laser measuring points but also reflects the degree of anomaly (such as the magnitude of frequency decrease and the factor of increase in damping ratio) through color intensity. It can also dynamically display whether multiple abnormal points are connected in a linear pattern (crack trend), a planar pattern (layered region), or exhibit gradient diffusion (material aging), thereby revealing the spatial evolution pattern of damage. In addition, the anomaly distribution map can be directly exported as maintenance work order coordinates, guiding maintenance personnel to accurately reach the designated location for ultrasonic or infrared re-inspection, significantly shortening downtime and providing spatial accuracy and decision support for the full life cycle health management of the target blade.

[0047] Through the above steps S102 to S110, the target vibration data of multiple laser measuring points can be determined by acquiring initial vibration data of multiple laser measuring points, position coordinates, attitude data and pointing data of multiple sampling periods, and feature extraction of the target vibration data to determine whether there are abnormal laser measuring points. When abnormal laser measuring points exist, the position of the abnormal measuring points on the target blade can be determined based on the position coordinates, attitude data and pointing data corresponding to the abnormal laser measuring points, thereby accurately obtaining the abnormal distribution map of the target blade. This achieves the technical effect of improving the abnormal positioning accuracy of wind turbine blades, and solves the technical problem of low abnormal positioning accuracy of wind turbine blades caused by UAV motion interference.

[0048] Based on the above embodiments and optional embodiments, the present invention proposes an optional implementation method. Figure 3 This is a flowchart of an optional anomaly detection method for wind turbine blades according to an embodiment of the present invention, such as... Figure 3 As shown, the method includes:

[0049] S1: System preparation and model import, specifically including: Before the UAV inspection, importing a high-precision 3D digital model of the target blade into the target blade anomaly detection system via the ground control station, i.e., a preset 3D model of the blade (including the target blade's geometric parameters, material properties, and theoretical modal characteristics), and establishing a health baseline database for the blade based on its historical vibration data, design parameters, or initial inspection results. This health baseline database should at least include the preset natural frequency and preset damping ratio of the target blade in a damage-free state. Based on the target blade's structural layout and environmental conditions, planning the UAV's hovering observation position and flight path, and setting the laser vibration measurement coverage area and measurement parameters (such as sampling rate and measurement point spacing).

[0050] S2: UAV positioning and quantum inertial navigation initialization, specifically including: the UAV takes off and autonomously flies to a preset hovering point (e.g., 30 to 40 meters in front of the impeller), ensuring that the laser vibration measurement beam can illuminate the blade surface without obstruction. The quantum inertial measurement unit is activated to complete high-precision initial alignment and record the UAV's position coordinates and attitude data for multiple sampling periods.

[0051] S3: The laser vibrometer simultaneously acquires initial vibration data corresponding to multiple laser measurement points. Specifically, this includes: starting the laser vibration measurement module according to the selected configuration mode. For example, in configuration mode A (single-point scanning), a two-dimensional turntable drives a miniature single-point laser vibrometer to scan laser measurement points one by one along the spanwise direction of the target blade, with a predetermined distance between adjacent laser measurement points and a preset dwell time at each laser measurement point; in configuration mode B (parallel array), a multi-channel laser vibration measurement array covers a predetermined width area of ​​the target blade at once, with all channels simultaneously acquiring initial vibration data without dwell time, performing continuous scanning. During the acquisition process, the laser vibrometer and quantum inertial navigation achieve time-synchronized triggering.

[0052] S4: Determine the target vibration data corresponding to each of the multiple laser measuring points. Specifically, this includes: separating the displacement projection data caused by the UAV motion interference from the initial vibration data corresponding to each of the multiple laser measuring points, thereby obtaining the target vibration data corresponding to each of the multiple laser measuring points. The specific implementation process is the same as in the aforementioned embodiment, and will not be repeated here.

[0053] S5: Perform feature extraction processing on the target vibration data corresponding to each of the multiple laser measuring points, and determine whether there are abnormal laser measuring points. Specifically, this includes: analyzing the target vibration data corresponding to each of the multiple laser measuring points, extracting the natural frequency and damping ratio corresponding to each of the multiple laser measuring points, and comparing them with the preset natural frequency and preset damping ratio in the health baseline database to determine whether there are abnormal laser measuring points. The specific implementation process is the same as in the aforementioned embodiment, and will not be repeated here.

[0054] S6: In the case of abnormal laser measurement points, determine the abnormal distribution map of the target blade. Specifically, if there are multiple abnormal laser measurement points, calculate the position coordinates of each of the multiple abnormal laser measurement points relative to the center of the wind turbine, i.e., the first position coordinates; map the first position coordinates corresponding to each of the multiple abnormal laser measurement points to the preset three-dimensional model of the blade to obtain the position coordinates of the multiple abnormal laser measurement points in the blade coordinate system, i.e., the second position coordinates, thereby obtaining the abnormal distribution map. The specific implementation process is the same as the aforementioned embodiment, and will not be repeated here.

[0055] This embodiment can achieve at least one of the following effects: (1) The first combination of quantum inertial measurement unit and laser vibration measurement. By utilizing the drift-free and high-precision characteristics of quantum inertial measurement unit, the motion separation problem in UAV-borne vibration measurement can be solved, and centimeter-level positioning of defects can be achieved at the same time; (2) Dual-mode laser vibration measurement scheme, providing two configurations: single-point scanning and array parallel, which are adapted to lightweight UAVs and heavy-duty UAVs respectively, taking into account different application scenarios; (3) Three-dimensional vibration field reconstruction: by multi-point synchronous measurement or scanning reconstruction, the complete vibration distribution on the blade surface can be obtained, which greatly improves the sensitivity of damage detection.

[0056] This embodiment also provides an anomaly detection device for wind turbine blades. This device is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the terms "module" and "device" can refer to a combination of software and / or hardware that performs a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0057] According to an embodiment of the present invention, an apparatus embodiment for implementing the above-described method for detecting anomalies in wind turbine blades is also provided. Figure 4 This is a schematic diagram of the structure of an abnormality detection device for wind turbine blades according to an embodiment of the present invention, as shown below. Figure 4 As shown, the above-mentioned anomaly detection device for wind turbine blades includes: a data acquisition module 400, a target vibration data determination module 402, a vibration characteristic data determination module 404, an anomaly laser measurement point judgment module 406, and an anomaly distribution map determination module 408, wherein:

[0058] The data acquisition module 400 is used to acquire the initial vibration data corresponding to multiple laser measuring points on the surface of the target blade of the wind turbine generator set collected in multiple sampling periods, the position coordinates and attitude data of the UAV corresponding to each of the multiple sampling periods, and the pointing data of the gimbal corresponding to each of the multiple sampling periods. The initial vibration data is obtained by sampling the corresponding laser measuring points along a preset inspection path using the laser vibration meter in the UAV. Multiple sampling periods correspond one-to-one with multiple laser measuring points. The position coordinates represent the position coordinates of the UAV relative to the center of the wind turbine generator set. The gimbal is a rotating platform installed in the UAV to control the sampling direction of the laser beam.

[0059] The target vibration data determination module 402 is connected to the data acquisition module 400. It is used to obtain the target vibration data corresponding to each of the multiple laser measuring points based on the initial vibration data corresponding to each of the multiple laser measuring points, as well as the position coordinates, attitude data and pointing data corresponding to each of the multiple sampling periods. The target vibration data represents the vibration data caused by the vibration of the target blade after eliminating the interference of the UAV motion.

[0060] The vibration feature data determination module 404 is connected to the target vibration data determination module 402 and is used to perform feature extraction processing on the target vibration data corresponding to multiple laser measuring points respectively to obtain the vibration feature data corresponding to multiple laser measuring points.

[0061] The abnormal laser measuring point judgment module 406 is connected to the vibration characteristic data determination module 404 and is used to determine whether there is an abnormal laser measuring point among the multiple laser measuring points based on the vibration characteristic data corresponding to each of the multiple laser measuring points and the preset health characteristic data.

[0062] The abnormal distribution map determination module 408 is connected to the abnormal laser measuring point judgment module 406. It is used to determine the abnormal distribution map of the target blade based on the position coordinates, attitude data and pointing data corresponding to the abnormal laser measuring points when there are abnormal laser measuring points among multiple laser measuring points.

[0063] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0064] It should be noted that the data acquisition module 400, target vibration data determination module 402, vibration characteristic data determination module 404, abnormal laser measurement point judgment module 406, and abnormal distribution map determination module 408 mentioned above correspond to steps S102 to S110 in the embodiments. The examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can run on a computer terminal.

[0065] It should be noted that the optional or preferred implementation methods of this embodiment can be found in the relevant descriptions in the embodiments, and will not be repeated here.

[0066] The aforementioned wind turbine blade anomaly detection device may also include a processor and a memory. The aforementioned data acquisition module 400, target vibration data determination module 402, vibration characteristic data determination module 404, abnormal laser measurement point judgment module 406, and abnormal distribution map determination module 408 are all stored in the memory as program modules. The processor executes the aforementioned program modules stored in the memory to realize the corresponding functions.

[0067] The processor contains a core that retrieves the corresponding program modules from memory. One or more cores may be configured. Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory includes at least one memory chip.

[0068] According to an embodiment of this application, an embodiment of a non-volatile storage medium is also provided. Optionally, in this embodiment, the non-volatile storage medium includes a stored program, wherein, when the program is running, it controls the device containing the non-volatile storage medium to execute any of the above-mentioned methods for detecting abnormalities in wind turbine blades.

[0069] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals, and the non-volatile storage medium includes stored programs.

[0070] Optionally, during program execution, the device containing the non-volatile storage medium is controlled to perform the following functions: acquire initial vibration data corresponding to multiple laser measuring points on the surface of the target blade of the wind turbine generator set collected during multiple sampling periods, position coordinates and attitude data of the UAV corresponding to each of the multiple sampling periods, and pointing data of the gimbal corresponding to each of the multiple sampling periods. The initial vibration data is obtained by sampling the corresponding laser measuring points along a preset inspection path using a laser vibration meter in the UAV. Multiple sampling periods correspond one-to-one with multiple laser measuring points. The position coordinates represent the position coordinates of the UAV relative to the center of the wind turbine rotor. The gimbal is a rotating platform installed in the UAV to control the laser beam sampling direction. Based on the initial vibration data corresponding to each of the multiple laser measuring points... The system uses dynamic data, along with position coordinates, attitude data, and pointing data corresponding to each of the multiple sampling periods, to obtain target vibration data for each of the multiple laser measuring points. The target vibration data represents the vibration data caused by the vibration of the target blade after excluding interference from UAV movement. Feature extraction processing is performed on the target vibration data corresponding to each of the multiple laser measuring points to obtain vibration feature data for each laser measuring point. Based on the vibration feature data corresponding to each of the multiple laser measuring points, and preset health feature data, it is determined whether there are any abnormal laser measuring points among the multiple laser measuring points. If abnormal laser measuring points are found, an abnormal distribution map of the target blade is determined based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measuring points.

[0071] According to an embodiment of this application, an embodiment of a processor is also provided. Optionally, in this embodiment, the processor is used to run a program, wherein the program executes any of the above-described methods for detecting abnormalities in wind turbine blades.

[0072] According to an embodiment of this application, an embodiment of a computer program product is also provided. Optionally, in this embodiment, the computer program product includes a computer program that, when executed by a processor, implements the steps of the above-described method for anomaly detection of wind turbine blades.

[0073] Optionally, when the aforementioned computer program product is executed on a data processing device, it is suitable to execute an initialization program with the following method steps: acquiring initial vibration data corresponding to multiple laser measuring points on the surface of the target blade of the wind turbine generator set collected during multiple sampling periods, position coordinates and attitude data of the UAV corresponding to each of the multiple sampling periods, and pointing data of the gimbal corresponding to each of the multiple sampling periods. The initial vibration data is obtained based on sampling of the corresponding laser measuring points by a laser vibrometer in the UAV along a preset inspection path. Multiple sampling periods correspond one-to-one with multiple laser measuring points. The position coordinates represent the position coordinates of the UAV relative to the center of the wind turbine rotor. The gimbal is a rotating platform installed in the UAV to control the laser beam sampling direction. Based on multiple laser measuring points... Based on the initial vibration data and the position coordinates, attitude data, and pointing data corresponding to each of the multiple sampling periods, target vibration data corresponding to multiple laser measuring points are obtained. The target vibration data represents the vibration data caused by the vibration of the target blade after excluding interference from UAV motion. Feature extraction processing is performed on the target vibration data corresponding to each of the multiple laser measuring points to obtain vibration feature data corresponding to each of the multiple laser measuring points. Based on the vibration feature data corresponding to each of the multiple laser measuring points and the preset health feature data, it is determined whether there are abnormal laser measuring points among the multiple laser measuring points. If there are abnormal laser measuring points among the multiple laser measuring points, the abnormal distribution map of the target blade is determined based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measuring points.

[0074] like Figure 5As shown, this embodiment of the invention provides an electronic device 10, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: acquiring initial vibration data corresponding to multiple laser measurement points on the surface of a target blade in a wind turbine generator set collected during multiple sampling periods; position coordinates and attitude data of a UAV corresponding to each of the multiple sampling periods; and pointing data of a gimbal corresponding to each of the multiple sampling periods. The initial vibration data is obtained based on sampling the corresponding laser measurement points along a preset inspection path using a laser vibration meter in the UAV. Multiple sampling periods correspond one-to-one with multiple laser measurement points. The position coordinates represent the position coordinates of the UAV relative to the center of the wind turbine rotor. The gimbal is a rotating device installed in the UAV to control the sampling direction of the laser beam. The process involves platform transformation; based on the initial vibration data corresponding to multiple laser measuring points, and the position coordinates, attitude data, and pointing data corresponding to multiple sampling periods, target vibration data corresponding to each laser measuring point is obtained. The target vibration data represents the vibration data caused by the vibration of the target blade after excluding interference from UAV movement. Feature extraction processing is performed on the target vibration data corresponding to each laser measuring point to obtain vibration feature data corresponding to each laser measuring point. Based on the vibration feature data corresponding to each laser measuring point and preset health feature data, it is determined whether there are any abnormal laser measuring points among the multiple laser measuring points. If abnormal laser measuring points are found, the abnormal distribution map of the target blade is determined based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measuring points.

[0075] The order of the above embodiments of the present invention is merely for description and does not represent the superiority or inferiority of the embodiments.

[0076] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0077] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of modules described above can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between modules, and may be electrical or other forms.

[0078] The modules described above as separate components may or may not be physically separate. Similarly, the components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple modules. Some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0079] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0080] If the aforementioned integrated modules are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable non-volatile storage medium. Based on this understanding, the technical solution of this invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a non-volatile storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned non-volatile storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0081] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for detecting anomalies in wind turbine blades, characterized in that, include: The method acquires initial vibration data corresponding to multiple laser measuring points on the surface of the target blade of the wind turbine generator set collected during multiple sampling periods, position coordinates and attitude data of the UAV corresponding to each of the multiple sampling periods, and pointing data of the gimbal corresponding to each of the multiple sampling periods. The initial vibration data is obtained by sampling the corresponding laser measuring points along a preset inspection path using a laser vibration meter in the UAV. The multiple sampling periods correspond one-to-one with the multiple laser measuring points. The position coordinates represent the position coordinates of the UAV relative to the center of the wind turbine generator set. The gimbal is a rotating platform installed in the UAV to control the sampling direction of the laser beam. Based on the initial vibration data corresponding to each of the plurality of laser measuring points, and the position coordinates, attitude data and pointing data corresponding to each of the plurality of sampling time periods, the target vibration data corresponding to each of the plurality of laser measuring points is obtained, wherein the target vibration data represents the vibration data caused by the vibration of the target blade after excluding the interference of the UAV's motion; Feature extraction processing is performed on the target vibration data corresponding to each of the plurality of laser measuring points to obtain vibration feature data corresponding to each of the plurality of laser measuring points; Based on the vibration characteristic data corresponding to each of the plurality of laser measuring points, and the preset health characteristic data, it is determined whether there are any abnormal laser measuring points among the plurality of laser measuring points; If an abnormal laser measuring point exists among the plurality of laser measuring points, an abnormal distribution map of the target blade is determined based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measuring point.

2. The method according to claim 1, characterized in that, The process of obtaining target vibration data corresponding to each of the plurality of laser measuring points based on the initial vibration data corresponding to each of the plurality of laser measuring points, and the position coordinates, attitude data, and pointing data corresponding to each of the plurality of sampling time periods, includes: Based on the attitude data and pointing data corresponding to each of the multiple sampling time periods, the laser beam pointing vector corresponding to each of the multiple laser measurement points is obtained, wherein the laser beam pointing vector is used to indicate the direction of the laser beam pointing to the corresponding laser measurement point in the corresponding sampling time period; Based on the position coordinates corresponding to each of the multiple sampling time periods and the laser beam pointing vector corresponding to each of the multiple laser measuring points, displacement projection data corresponding to each of the multiple laser measuring points is obtained, wherein the displacement projection data represents the interference vibration data obtained by the movement of the UAV at the corresponding laser measuring point; Based on the initial vibration data and displacement projection data corresponding to each of the plurality of laser measuring points, the target vibration data corresponding to each of the plurality of laser measuring points is obtained.

3. The method according to claim 1, characterized in that, When the vibration feature data includes the natural frequency and damping ratio of the corresponding laser measuring point, the step of performing feature extraction processing on the target vibration data corresponding to each of the plurality of laser measuring points to obtain the vibration feature data corresponding to each of the plurality of laser measuring points includes: The target vibration data corresponding to each of the plurality of laser measuring points are denoised to obtain the denoised vibration data corresponding to each of the plurality of laser measuring points. The denoised vibration data corresponding to each of the plurality of laser measuring points are subjected to time-frequency domain transformation processing to obtain the time-frequency energy distribution map corresponding to each of the plurality of laser measuring points. The time-frequency domain transformation is a short-time Fourier transform or a wavelet transform. The time-frequency energy distribution map is used to indicate the vibration energy distribution characteristics of the corresponding laser measuring point in the time-frequency domain. Based on the time-frequency energy distribution map corresponding to each of the plurality of laser measuring points, the natural frequency corresponding to each of the plurality of laser measuring points is obtained, wherein the natural frequency represents the resonance frequency of the target blade at the corresponding laser measuring point; The denoised vibration data corresponding to each of the plurality of laser measuring points are subjected to Hilbert transform processing to obtain the amplitude envelope diagrams corresponding to each of the plurality of laser measuring points. The horizontal axis of the amplitude envelope diagram represents the multiple sampling times in the sampling period of the corresponding laser measuring point, and the vertical axis of the amplitude envelope diagram represents the vibration displacement amplitude of the corresponding laser measuring point at each of the multiple sampling times. Based on the amplitude envelope diagrams corresponding to the plurality of laser measuring points, the damping ratios corresponding to the plurality of laser measuring points are obtained, wherein the damping ratios are used to quantify the dissipation rate of the vibration energy of the target blade at the corresponding laser measuring point.

4. The method according to claim 3, characterized in that, The process of obtaining the natural frequencies corresponding to each of the plurality of laser measuring points based on the time-frequency energy distribution maps corresponding to each of the plurality of laser measuring points includes: The vibration energy amplitude that is greater than a preset first amplitude threshold in the time-frequency energy distribution map of any one of the plurality of laser measuring points is taken as the target vibration energy amplitude of any one of the laser measuring points. The frequency corresponding to the amplitude of the target vibration energy is determined to be the natural frequency of any laser measuring point; The natural frequencies corresponding to each of the plurality of laser measuring points are obtained by using the method of obtaining the natural frequency of any one of the laser measuring points.

5. The method according to claim 3, characterized in that, The step of obtaining the damping ratio corresponding to each of the plurality of laser measuring points based on the amplitude envelope diagrams corresponding to each of the plurality of laser measuring points includes: The vibration displacement amplitude that is greater than a preset second amplitude threshold in the amplitude envelope of any laser measuring point among the plurality of laser measuring points is taken as the target vibration displacement amplitude of any laser measuring point. Based on the target vibration displacement amplitude and the vibration displacement amplitude at a predetermined sampling time after the target vibration displacement amplitude, the damping ratio of any laser measuring point is obtained; The damping ratio corresponding to each of the plurality of laser measuring points is obtained by using the method of obtaining the damping ratio of any one of the laser measuring points.

6. The method according to claim 1, characterized in that, When the vibration characteristic data includes natural frequency and damping ratio, and the preset health characteristic data includes preset natural frequency and preset damping ratio, the step of determining whether there are abnormal laser measuring points among the plurality of laser measuring points based on the vibration characteristic data corresponding to each of the plurality of laser measuring points and the preset health characteristic data includes: Based on the natural frequency and the preset natural frequency corresponding to each of the plurality of laser measuring points, the natural frequency deviation corresponding to each of the plurality of laser measuring points is obtained; Based on the damping ratio and the preset damping ratio corresponding to each of the plurality of laser measuring points, the damping ratio deviation corresponding to each of the plurality of laser measuring points is obtained; If, among the plurality of laser measuring points, the natural frequency deviation of any laser measuring point is greater than a preset frequency deviation, or the damping ratio deviation of any laser measuring point is greater than a preset damping ratio deviation, then that laser measuring point is determined to be the abnormal laser measuring point; or If the inherent frequency deviation corresponding to each of the plurality of laser measuring points is less than or equal to the preset frequency deviation, or if the damping ratio deviation corresponding to each of the plurality of laser measuring points is less than or equal to the preset damping ratio deviation, it is determined that there is no abnormal laser measuring point among the plurality of laser measuring points.

7. The method according to any one of claims 1 to 6, characterized in that, When there are multiple abnormal laser measurement points, and when the abnormal laser measurement point exists among the multiple laser measurement points, the abnormal distribution map of the target blade is determined based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measurement point, including: Based on the target distance, position coordinates, attitude data, and pointing data corresponding to each of the multiple abnormal laser measuring points, the first position coordinates corresponding to each of the multiple abnormal laser measuring points are obtained, wherein the target distance represents the straight-line distance between the corresponding abnormal laser measuring point and the UAV, and the first position coordinates represent the position coordinates of the corresponding abnormal laser measuring point relative to the center of the wind turbine. The first position coordinates corresponding to each of the plurality of abnormal laser measuring points are mapped onto a preset blade three-dimensional model to obtain the second position coordinates corresponding to each of the plurality of abnormal laser measuring points. The preset blade three-dimensional model represents a three-dimensional geometric model constructed based on the parameters of the target blade, and is used to indicate the mapping relationship between the coordinate system corresponding to the wind turbine center and the blade coordinate system corresponding to the target blade. The anomaly distribution map is obtained based on the second position coordinates corresponding to each of the multiple abnormal laser measurement points.

8. An anomaly detection device for wind turbine blades, characterized in that, include: The data acquisition module is used to acquire initial vibration data corresponding to multiple laser measuring points on the surface of the target blade of the wind turbine generator set collected during multiple sampling periods, position coordinates and attitude data of the UAV corresponding to each of the multiple sampling periods, and pointing data of the gimbal corresponding to each of the multiple sampling periods. The initial vibration data is obtained by sampling the corresponding laser measuring points along a preset inspection path using a laser vibration meter in the UAV. The multiple sampling periods correspond one-to-one with the multiple laser measuring points. The position coordinates represent the position coordinates of the UAV relative to the center of the wind turbine rotor of the wind turbine generator set. The gimbal is a rotating platform installed in the UAV for controlling the sampling direction of the laser beam. The target vibration data determination module is used to obtain target vibration data corresponding to each of the multiple laser measuring points based on the initial vibration data corresponding to each of the multiple laser measuring points, as well as the position coordinates, attitude data and pointing data corresponding to each of the multiple sampling time periods. The target vibration data represents the vibration data caused by the vibration of the target blade after the motion interference of the UAV is eliminated. The vibration feature data determination module is used to perform feature extraction processing on the target vibration data corresponding to each of the plurality of laser measuring points to obtain the vibration feature data corresponding to each of the plurality of laser measuring points. An abnormal laser measuring point judgment module is used to determine whether there is an abnormal laser measuring point among the multiple laser measuring points based on the vibration characteristic data corresponding to each of the multiple laser measuring points and preset health characteristic data. The anomaly distribution map determination module is used to determine the anomaly distribution map of the target blade based on the position coordinates, attitude data, and pointing data corresponding to the abnormal laser measuring points when the abnormal laser measuring points exist among the plurality of laser measuring points.

9. A non-volatile storage medium, characterized in that, The non-volatile storage medium stores multiple instructions, which are adapted to be loaded by a processor and executed by the anomaly detection method for wind turbine blades according to any one of claims 1 to 7.

10. An electronic device, characterized in that, The device includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method for detecting anomalies in wind turbine blades as described in any one of claims 1 to 7.

Images