Wind power single pile local scouring real-time detection method based on multi-frequency acoustic tomography

Abstract

The invention relates to the technical field of ocean engineering, in particular to a wind power single pile local scouring real-time detection method based on multi-frequency acoustic tomography, which comprises the following steps: S1, annularly arranging an acoustic sensor array on the section above the water surface of a pile body; s2, transmitting multi-frequency sound wave signals including low-frequency penetrating waves and high-frequency surface waves; s3, scattered wave signals reflected by the pile-seabed interface are collected, and low-frequency scattered waves and high-frequency scattered waves corresponding to the frequency in S2 are separated; s4, constructing a mud layer penetration acoustic tomography matrix and a surface topography acoustic tomography matrix; s5, three-dimensional data fusion is carried out, and a three-dimensional image of the scouring state of the seabed around the pile is generated; and S6, a scouring pit is positioned according to a sound wave propagation time delay sudden change area in the three-dimensional image of the scouring state. According to the method, the scouring state of the seabed around the wind power single pile foundation can be accurately monitored by combining the multi-frequency acoustic tomography technology and time delay abrupt change area detection.

Description

Technical Field

[0001] This invention relates to the field of marine engineering technology, and in particular to a real-time detection method for local scour of wind turbine monopiles based on multi-frequency acoustic tomography. Background Technology

[0002] In recent years, with the rapid development of the wind power industry, the construction scale of offshore wind power has continued to expand. In the monopile foundation structure of offshore wind power, the scouring phenomenon of the seabed around the piles poses a significant threat to the stability and safety of the wind turbine. Seabed scouring can lead to the exposure of the pile foundation and a reduction in stability, thereby affecting the long-term operation and economic efficiency of the wind turbine.

[0003] Existing technologies still have certain limitations, especially in complex environments where the time delay changes of sound wave propagation are difficult to analyze accurately in real time, and scour information at different depths and locations is difficult to comprehensively display. Current methods typically cannot effectively combine the penetration of deep mud layers with the surface condition of the seabed for analysis. Therefore, there is an urgent need for a real-time detection method for localized scour of wind turbine monopiles based on multi-frequency acoustic tomography to address these issues. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides a real-time detection method for local scour of wind turbine monopile based on multi-frequency acoustic tomography.

[0005] A real-time detection method for local scour of wind turbine monopile based on multi-frequency acoustic tomography includes the following steps: S1: An acoustic sensor array is arranged in a ring above the water surface of the pile body. The acoustic sensor array includes at least 3 sets of orthogonally distributed transmitting-receiving units. S2: Transmits multi-frequency acoustic signals, including low-frequency penetrating waves and high-frequency surface waves, to the pile-seabed coupling area through the transmitting unit group; S3: The scattered wave signal reflected from the pile-seabed interface is collected by the receiving unit group, and the low-frequency scattered wave and high-frequency scattered wave corresponding to the S2 frequency are separated from the scattered wave signal by the time-frequency separation algorithm. S4: Construct a mud layer penetration acoustic tomography matrix based on the acoustic wave propagation delay distribution of low-frequency scattered waves, and simultaneously construct a surface morphology acoustic tomography matrix based on high-frequency scattered waves. S5: The acoustic tomography matrix of mud penetration is fused with the acoustic tomography matrix of surface morphology to generate a three-dimensional image of the scour state of the seabed around the pile. S6: Locate the scour pit based on the abrupt change in acoustic wave propagation time delay in the 3D image of the scour state.

[0006] Optionally, S1 specifically includes: S11: N installation positions are set at equal intervals along the circumference of the pile body. Each installation position is fixed with an adjustable universal bracket, and a sensor base is integrated at the end of the bracket. S12: Install at least 3 sets of transmit-receive unit groups on the sensor base in triaxial orthogonal directions of 0°, 45° and 90°; S13: Measure the curvature of the pile surface using a laser rangefinder and adjust the bracket angle so that the axis of the sensor unit group always matches the normal direction of the pile surface within a predetermined range.

[0007] Optionally, S2 specifically includes: S21: Let the center frequency of the low-frequency penetrating wave be... Its frequency range is Let the center frequency of the high-frequency surface wave be... Its frequency range is ;in ; S22: Within a set single transmission cycle T, low-frequency penetrating waves are transmitted during the first 0.3T period; high-frequency surface waves are transmitted during the last 0.7T period. S23: Through the phased array of the transmitting unit group, low-frequency penetrating waves are focused on the mud layer infiltration zone at a depth of 0-5m around the pile, while high-frequency surface waves cover the seabed surface area within a range of 0-2m around the pile.

[0008] Optionally, S3 specifically includes: S31: The scattered wave signal reflected from the pile-seabed interface is collected by the receiving unit group. The scattered wave signal includes reflection components from low-frequency penetrating waves and high-frequency surface waves.

[0009] S32: The time-frequency separation algorithm is used to process the acquired scattered wave signal. Specifically, the scattered wave signal is first decomposed according to the characteristics of the time domain and frequency domain; then, the low-frequency scattered wave and high-frequency scattered wave corresponding to the S2 frequency are identified and extracted.

[0010] Optionally, the decomposition of the scattered wave signal according to its time-domain and frequency-domain characteristics includes: S321: Based on the short-time Fourier transform, the collected scattered wave signal is divided into time windows, and the Fourier transform is performed on the signal in each time window to obtain the frequency components in the corresponding time period. S322: Based on the Fourier transform, the time-domain signal is converted into a frequency-domain signal; by integrating the time-domain signal, the signal components in the frequency domain are obtained.

[0011] Optionally, identifying and extracting the low-frequency and high-frequency scattered waves corresponding to the S2 frequency includes: Signal frequency range division: Based on the frequency range of the low-frequency and high-frequency signals generated in S2, the frequency intervals of the low-frequency and high-frequency signals are set; Frequency filtering: Using low-pass and high-pass filters to process frequency domain signals. Filtering is performed to separate the low-frequency components. and high-frequency components ; Signal reconstruction: Reconstructing the filtered low-frequency components and high-frequency components The inverse Fourier transform is performed back to the time domain signal to obtain the corresponding low-frequency and high-frequency time domain signals, and finally the low-frequency and high-frequency scattered wave signals are obtained.

[0012] Optionally, constructing the mud-penetrating acoustic tomography matrix includes: Sound wave path calculation: Based on the time delay distribution data of low-frequency scattered waves, combined with the sound velocity information of seawater and silt, the propagation path of sound waves from the emission point to the receiving point is calculated by the curved ray tracing inversion algorithm, and the specific coordinates of each path are determined. Path correction: Based on the measured flow velocity and seawater viscosity, the radius of curvature of the sound wave path is adjusted using the Reynolds number to correct the propagation characteristics of the path and compensate for the influence of fluid dynamics on sound wave propagation. Path Discretization and Attenuation Rate Calculation: The corrected acoustic path is discretized into a voxel grid, and the energy attenuation rate of each path is calculated. A mud penetration acoustic tomography matrix is ​​generated using the ratio of the transmitted and received wave amplitudes. .

[0013] Optionally, the construction of the surface topography acoustic tomography matrix based on high-frequency scattered waves includes: Finite-difference time-domain beamforming: The wave field intensity of each three-dimensional grid point is calculated by using the finite-difference time-domain method. The wave field intensity of the grid point is obtained by multiplying the signal of each receiving unit group by its weight after the propagation time correction, and then summing the contributions of all receiving units by weight. Wave field intensity extraction and normalization: Extract wave field intensity data of the seabed surface layer and normalize it to the [0,1] interval; Generate the acoustic tomography matrix: Construct the surface topography acoustic tomography matrix using normalized wave field intensity data. .

[0014] Optionally, S5 specifically includes: S51: Obtain the acoustic tomography matrix of mud penetration constructed in S4 and surface morphology acoustic tomography matrix This ensures that the two are aligned in spatial coordinates; S52: Yes and The elements in the array are standardized to ensure that their numerical ranges are consistent. S53: Using a weighted average method, the standardized... and The data are fused, and the resulting 3D image of the scour state is calculated. The expression is: ,in, and These are the weighting coefficients for the mud layer and the seabed surface layer, respectively. The generated 3D image of the scouring state; S54: Merge the matrix Converted into a three-dimensional scour state image.

[0015] Optionally, S6 specifically includes: S61: Extract acoustic wave propagation delay data from the 3D image of the scour state generated by S5, and identify the regions where the acoustic wave propagation delay changes significantly by setting a delay abrupt change threshold, i.e., delay abrupt change regions. S62: After identifying the time delay abrupt change region, locate its spatial coordinates, determine the boundary of the abrupt change region, and compare it with known seabed morphology data to locate the specific scour pit location. S63: Extract the outline of the abrupt change region to determine the shape and size of the scour pit.

[0016] The beneficial effects of this invention are: This invention, by combining the different characteristics of low-frequency scattered waves and high-frequency surface waves, can acquire detailed scour information at different depths and locations, accurately depicting the scour state of the seabed. This technology can effectively overcome the limitations of traditional monitoring methods, providing high-resolution three-dimensional images of the scour state and greatly improving the accuracy of scour monitoring.

[0017] This invention achieves precise location and shape recognition of scour pits through time-delay abrupt change region detection and contour extraction; this not only enables the seabed scour situation to be reflected in real time, but also provides reliable data support for subsequent risk assessment and decision-making. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of a real-time detection method for local scour of a single wind turbine pile according to an embodiment of the present invention; Figure 2This is a schematic diagram of the process of deploying an acoustic sensor array according to an embodiment of the present invention. Detailed Implementation

[0020] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0021] It should be noted that the use of terms such as "an embodiment," "an embodiment," "an exemplary embodiment," and "some embodiments" in the specification indicates that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the art.

[0022] In general, terms can be understood, at least in part, from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending at least in part on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0023] like Figure 1-Figure 2 As shown, a real-time detection method for local scour of a wind turbine monopile based on multi-frequency acoustic tomography includes the following steps: S1: An acoustic sensor array is arranged in a ring above the water surface of the pile body. The acoustic sensor array includes at least 3 sets of orthogonally distributed transmitting-receiving units. S1 specifically includes: S11: N installation positions (N≥6) are set at equal intervals along the circumference of the pile body. Each installation position is fixed with an adjustable universal bracket, and a sensor base is integrated at the end of the bracket. S12: At least three sets of transmit-receive unit groups are installed on the sensor base in orthogonal directions of 0°, 45°, and 90°. The axis of the 0° unit group is perpendicular to the normal direction of the pile surface; the axis of the 45° unit group forms a 45° angle with the normal direction of the pile surface; and the axis of the 90° unit group is parallel to the axis of the pile. S13: The curvature of the pile surface is measured using a laser rangefinder, and the bracket angle is adjusted to ensure that the axis of the sensor unit group is always matched with the normal direction of the pile surface within a predetermined range, thereby ensuring that the sensor acquires signals in the optimal position; the relationship between the radius of curvature of the pile surface and the pitch angle of the universal bracket is as follows: Where R is the radius of curvature of the pile surface; d represents the pitch angle of the universal joint; d is the distance between the probe end face of the unit group and the surface of the pile, controlling... .

[0024] S2: Transmits multi-frequency acoustic signals, including low-frequency penetrating waves and high-frequency surface waves, to the pile-seabed coupling area through the transmitting unit group; S2 specifically includes: S21: Let the center frequency of the low-frequency penetrating wave be... Its frequency range is Let the center frequency of the high-frequency surface wave be... Its frequency range is ;in ; S22: Within a set single transmission cycle T ( It emits low-frequency penetrating waves during the first 0.3T period and high-frequency surface waves during the second 0.7T period. S23: Through the phased array of the transmitting unit group, low-frequency penetrating waves are focused on the mud layer infiltration zone at a depth of 0-5m around the pile, while high-frequency surface waves cover the seabed surface area within a range of 0-2m around the pile. By precisely controlling the transmission timing and frequency range of the dual-frequency signals, and combining it with the directional beamforming of the phased array, efficient acoustic detection of the pile-seabed coupling area can be achieved, enhancing the penetration capability of mud layers and seabed surface areas at different depths, thereby improving the accuracy and effectiveness of real-time monitoring.

[0025] S3: The scattered wave signal reflected from the pile-seabed interface is collected by the receiving unit group, and the low-frequency scattered wave and high-frequency scattered wave corresponding to the S2 frequency are separated from the scattered wave signal by the time-frequency separation algorithm. S3 specifically includes: S31: The scattered wave signal reflected from the pile-seabed interface is collected by the receiving unit group. The scattered wave signal includes the reflection components from low-frequency penetrating waves and high-frequency surface waves. The signal is collected in parallel by multiple receiving unit groups to ensure high-resolution time and space information.

[0026] S32: The collected scattered wave signal is processed using a time-frequency separation algorithm. Specifically, the scattered wave signal is first decomposed according to its time and frequency domain characteristics; then, low-frequency and high-frequency scattered waves corresponding to the S2 frequency are identified and extracted. By using the time-frequency separation algorithm to separate the scattered wave signal, low-frequency and high-frequency scattered waves can be effectively extracted, thereby accurately distinguishing information between the mud layer infiltration zone and the seabed surface zone. This technology ensures the separation and analysis of signals of different frequencies, enhances the accuracy and reliability of signal processing, and provides more accurate raw data for subsequent three-dimensional image generation.

[0027] The scattered wave signal is decomposed according to its characteristics in the time and frequency domains, including: S321: Based on the Short-Time Fourier Transform (STFT), the acquired scattered wave signal is divided into time windows, and a Fourier Transform is performed on the signal within each time window to obtain the frequency components within the corresponding time period. This method allows us to obtain the signal's distribution in time and frequency, thereby analyzing the frequency characteristics within different time periods. The formula for the Short-Time Fourier Transform is: ,in, The signal components in the time-frequency domain are represented by f, where f is the frequency and t is the time. Indicates the original signal in time The value on; It is a window function used to limit the time range of signal analysis, where t is the center position of the current window and f is the frequency; is the time variable; j is the imaginary unit; through this formula, the signal is decomposed into a joint representation with time and frequency components, which can be used for subsequent frequency domain analysis; S322: According to the Fourier transform, the time-domain signal is converted into a frequency-domain signal; by integrating the time-domain signal, the signal components in the frequency domain are obtained; by analyzing the frequency distribution of the signal, signals of different frequencies can be distinguished, thus providing a basis for subsequent separation of low-frequency and high-frequency signals; the formula for the Fourier transform is as follows: ,in, For frequency domain signals, it represents the signal strength at frequency f; f is the time-domain signal, representing the signal value over time t; f is the frequency; t is the time; j is the imaginary unit; the Fourier transform extracts the frequency component by integrating the signal's value over the entire time range, thus obtaining the signal representation in the frequency domain.

[0028] Identifying and extracting the low-frequency and high-frequency scattered waves corresponding to the S2 frequency includes: Signal frequency range division: Based on the frequency ranges of the low-frequency and high-frequency signals generated in S2, the frequency intervals of the low-frequency and high-frequency signals are defined. Specifically, the frequency range of the low-frequency scattered wave is defined as follows: The frequency range of high-frequency scattered waves is defined as and determine the frequency threshold. Used for signal differentiation, where: ; Frequency filtering: Using low-pass and high-pass filters to process frequency domain signals. Filtering is performed to separate the low-frequency components. and high-frequency components The low-frequency portion is filtered by a low-pass filter. The filtering expression is as follows: The high-frequency portion is filtered by a high-pass filter. The filtering expression is as follows: ,in, and These are filter functions for low and high frequencies, respectively, used to control the frequency range of the signal; For low-frequency filters, it uses the form of a low-pass filter, below the threshold frequency. The region through which the signal passes is attenuated for frequencies exceeding this range; the expression for this is: ; For high-frequency filters, it is typically used in the form of a high-pass filter, which is applied above the threshold frequency. The region where the signal passes through is attenuated for frequencies below this value; its expression is: ; Signal reconstruction: Reconstructing the filtered low-frequency components and high-frequency components The inverse Fourier transform is performed back to the time domain signal to obtain the corresponding low-frequency and high-frequency time domain signals, ultimately yielding the low-frequency and high-frequency scattered wave signals. By applying these two filter functions, the frequency domain signal can be effectively divided into low-frequency and high-frequency components, thereby accurately extracting the scattered wave signals of the corresponding frequencies. This separation method ensures accurate signal extraction and enhances the reliability and accuracy of the analysis. In particular, in complex seabed environments, it helps to improve the accuracy and effectiveness of local scour detection.

[0029] S4: Construct a mud layer penetration acoustic tomography matrix based on the acoustic wave propagation delay distribution of low-frequency scattered waves, and simultaneously construct a surface morphology acoustic tomography matrix based on high-frequency scattered waves. Constructing the acoustic tomography matrix through the mud layer includes: Sound wave path calculation: Based on the time delay distribution data of low-frequency scattered waves, combined with the sound velocity information of seawater and silt, the propagation path of the sound wave from the emission point to the receiving point is calculated using a curved ray tracing inversion algorithm, and the specific coordinates of each path are determined; the calculation formula is: ,in, The coordinates of the launch point; The speed of sound in seawater (unit: m / s, taken as 1500 m / s). The sound velocity in the silt (unit: m / s, measured in real time by a sound velocity meter). This is time-delay distribution data; The azimuth angle of the launch unit group; Let be the coordinates of the i-th path point; Path correction: Based on the measured flow velocity and seawater viscosity, the radius of curvature of the sound wave path is adjusted using the Reynolds number to correct the propagation characteristics of the path and compensate for the influence of fluid dynamics on sound wave propagation; specifically, let the Reynolds number be... Its expression is: Where U is the measured value from the flow velocity sensor, D is the pile diameter, and v is the kinematic viscosity of seawater; the formula for adjusting the radius of curvature of the correction path based on the Reynolds number is: ,in, Uncorrected radius of curvature; The corrected path curvature radius; Path Discretization and Attenuation Rate Calculation: The corrected acoustic path is discretized into a voxel grid, and the energy attenuation rate of each path is calculated. A mud penetration acoustic tomography matrix is ​​generated using the ratio of the transmitted and received wave amplitudes. This reveals the sound wave propagation characteristics of the mud layer, and the formula for calculating the energy attenuation rate is: ,in, The attenuation rate of sound wave energy; The amplitude of the transmitted wave; To receive the wave amplitude.

[0030] Constructing surface topography acoustic tomography matrices based on high-frequency scattered waves includes: Finite-Difference Time-Domain Beamforming: The wave field intensity at each three-dimensional grid point is calculated using the finite-difference time-domain method. This is achieved by multiplying the signal of each receiver group by its weight after propagation time correction, and then weighted summing the contributions of all receivers to obtain the wave field intensity at each grid point. The calculation formula is as follows: ,in, The weight of the k-th receiving unit; To receive signals; For sound waves from point The propagation time to the k-th receiving unit; K is the total number of receiving units; The wave field intensity; Wave field intensity extraction and normalization: Wave field intensity data of the seabed surface layer were extracted and normalized to the [0,1] interval to ensure data consistency for subsequent analysis; Generate the acoustic tomography matrix: Construct the surface topography acoustic tomography matrix using normalized wave field intensity data. This provides accurate acoustic features of the seabed surface. The above steps, through time-domain finite-difference beamforming and wave field intensity calculation, can effectively extract the acoustic features of the seabed surface. The normalized surface morphology acoustic tomography matrix provides standardized signal data for detailed analysis of the seabed surface.

[0031] S5: The acoustic tomography matrix of mud penetration is fused with the acoustic tomography matrix of surface morphology to generate a three-dimensional image of the scour state of the seabed around the pile. S5 specifically includes: S51: Obtain the acoustic tomography matrix of mud penetration constructed in S4 and surface morphology acoustic tomography matrix This ensures that the two are aligned in spatial coordinates, allowing them to be merged within the same three-dimensional spatial grid; S52: Yes and The elements in the array are standardized to ensure that their numerical ranges are consistent. S53: Using a weighted average method, the standardized... and The data are fused, and the resulting 3D image of the scour state is calculated. The expression is: ,in, and These are the weighting coefficients for the mud layer and the seabed surface layer, respectively. The generated 3D image of the scouring state; S54: Merge the matrix The data is converted into a three-dimensional scour state image to display the scour conditions of the seabed at different depths and on the surface around the pile. The above steps, by fusing three-dimensional data from the mud layer penetration acoustic tomography matrix and the surface morphology acoustic tomography matrix, can comprehensively reflect the scour state of the seabed around the pile and provide high-resolution scour images, providing accurate data support for subsequent risk assessment and maintenance decisions.

[0032] S6: Locate the scour pit based on the abrupt changes in acoustic wave propagation time delay in the 3D image of the scour state; S6 specifically includes: S61: Extract acoustic wave propagation delay data from the 3D image of the scour state generated by S5, and identify the regions where the acoustic wave propagation delay changes significantly by setting a delay abrupt change threshold, i.e., delay abrupt change regions. S62: After identifying the time delay abrupt change region, locate its spatial coordinates, determine the boundary of the abrupt change region, and compare it with known seabed morphology data to locate the specific scour pit location. S63: Extract the contour of the abrupt change region and determine the shape and size of the scour pit; through the above steps, the time delay abrupt change region can be accurately extracted from the three-dimensional image, and the location and contour of the scour pit can be accurately located. This process helps to monitor and analyze the scour condition of the seabed around the pile with high precision, and improves the accuracy and reliability of local scour detection.

[0033] Example 1: Real-time detection of wind turbine monopile scour in a coastal wind farm.

[0034] In order to ensure the structural stability of the wind turbine pile foundation under strong tidal scour conditions at a coastal offshore wind farm, the local scour real-time detection method based on multi-frequency acoustic tomography described in this invention was used to conduct on-site testing of the wind turbine monopile foundation numbered NT-07.

[0035] I. Deployment of Acoustic Sensor Array: At a height of 0.5 m above the water surface on the NT-07 monopile foundation, 12 equally spaced installation points were arranged around the pile, with each point equipped with a universal bracket and sensor base. Three sets of transmitter-receiver units were deployed, with azimuth angles of 0° (horizontal radial), 45° (oblique incidence), and 90° (axial). Using laser ranging, the pile radius was determined to be 1.8 m, and the probe was positioned 8 mm from the surface. The curvature adaptive formula was applied. ; Adjust the bracket's pitch angle to ensure that the three sets of sensors are aligned with the target area.

[0036] II. Multi-frequency acoustic wave emission: The transmission period is set to T=80ms, with the first 24ms transmitting a low-frequency penetrating wave (center frequency 20 kHz, bandwidth 10 kHz) and the last 56ms transmitting a high-frequency surface wave (center frequency 200 kHz, bandwidth 50 kHz). The phased array beamforming strategy enables: The low-frequency wave is focused on the mud layer with a radius of 2m and a depth of 0–5m around the pile. High-frequency waves cover the surface seabed with a radius of 2m and a depth of 0–2m around the pile.

[0037] III. Acquisition and Separation of Scattered Wave Signals: The receiving unit acquires the composite scattered wave signal reflected back from the pile-seabed interface in real time. The signal is then subjected to a short-time Fourier transform. ; Using a bandpass filter, low-frequency scattered waves in the 15-30kHz range and high-frequency scattered waves in the 100-300kHz range are extracted, and then reconstructed by inverse transform: Low-frequency signals; High-frequency signal.

[0038] IV. Constructing the tomography matrix: 4.1 Acoustic Tomography Matrix of Mud Layer Penetration : Utilizing low-frequency delay data Speed ​​of sound of seawater m / s, velocity of sound in silt m / s, calculate path points: ; Measured flow rate m / s, pile diameter D=3.6m, seawater kinematic viscosity Calculate the Reynolds number: ; Corrected radius of curvature: ; Calculate the acoustic energy attenuation rate per voxel: .

[0039] 4.2 Surface morphology acoustic tomography matrix : The wave field intensity at each three-dimensional grid point was calculated using beamforming: ; The wave field intensity at z=0 layer (seabed surface) is extracted and normalized to [0,1] to form... .

[0040] V. Fusing and Generating a 3D Image: Will and Spatially align and normalize on the grid. Let the fusion weight coefficient be... , Fusion formula: ; Fusion Matrix Converted into a 3D image, it shows the erosion intensity distribution of the seabed surface and mud layer.

[0041] VI. Identification and Location of Scour Pit: 6.1 Identification of Time Delay Abrupt Regions: From Extract the propagation delay distribution and calculate the difference between adjacent voxels: ; Set threshold ms, regions that meet the criteria are marked as mutation regions.

[0042] 6.2 Scour Pit Location: The three-dimensional boundary coordinates of the abrupt change area were extracted and compared with the original seabed morphology model to determine the center location of the scour pit as follows: ; 6.3 Contour Recognition and Size Determination: The extent of the scour pit was identified using the three-dimensional boundary growth method, and its approximate volume was calculated as follows: .

[0043] This embodiment demonstrates how the present invention enables full-process, quantifiable scour status detection and risk warning in a real offshore wind turbine foundation project. The 3D image generated by fusing low-frequency mud penetration and high-frequency surface imaging data accurately reconstructs the geometry and location of scour pits, providing clear data support for maintenance units to formulate reinforcement and backfilling measures.

[0044] The following describes the effects of this embodiment in conjunction with the table.

[0045] Table 1. Comparison of the accuracy of scour pit identification between the method of this embodiment and traditional sonar detection.

[0046] As can be seen from the data in Table 1, the method of this invention closely matches the actual situation in identifying the depth of scour pits, with a maximum relative error of less than 1.5%. In contrast, traditional single-frequency sonar methods, limited to surface detection, generally exhibit significant depth errors, with a maximum relative error exceeding 18%. This demonstrates that the present invention has a significant advantage in the spatial identification accuracy of scour pits.

[0047] Table 2 Comparison of response time between the method of this embodiment and other detection schemes for identifying scouring conditions.

[0048] Table 2 shows the differences in processing efficiency among different methods under the same data volume at the scour site. This invention simultaneously acquires features of the mud layer and seabed using multi-frequency acoustic signals and fuses them to generate a three-dimensional image, omitting many intermediate steps and keeping the total response time within 10 seconds. This significantly improves detection efficiency compared to traditional sonar and video image methods, making it suitable for real-time monitoring applications.

[0049] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0050] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A real-time detection method for local scour of a wind turbine monopile based on multi-frequency acoustic tomography, characterized in that, Includes the following steps: S1: An acoustic sensor array is arranged in a ring above the water surface of the pile body. The acoustic sensor array includes at least 3 sets of orthogonally distributed transmitting-receiving units. S2: Transmits multi-frequency acoustic signals, including low-frequency penetrating waves and high-frequency surface waves, to the pile-seabed coupling area through the transmitting unit group; S3: The scattered wave signal reflected from the pile-seabed interface is collected by the receiving unit group, and the low-frequency scattered wave and high-frequency scattered wave corresponding to the S2 frequency are separated from the scattered wave signal by the time-frequency separation algorithm. S4: Construct a mud layer penetration acoustic tomography matrix based on the acoustic wave propagation delay distribution of low-frequency scattered waves, and simultaneously construct a surface morphology acoustic tomography matrix based on high-frequency scattered waves. S5: The acoustic tomography matrix of mud penetration is fused with the acoustic tomography matrix of surface morphology to generate a three-dimensional image of the scour state of the seabed around the pile. S6: Locate the scour pit based on the abrupt change in acoustic wave propagation time delay in the 3D image of the scour state.

2. The method for real-time detection of local scour of a wind turbine monopile based on multi-frequency acoustic tomography according to claim 1, characterized in that, S1 specifically includes: S11: N installation positions are set at equal intervals along the circumference of the pile body. Each installation position is fixed with an adjustable universal bracket, and a sensor base is integrated at the end of the bracket. S12: Install at least 3 sets of transmit-receive unit groups on the sensor base in triaxial orthogonal directions of 0°, 45° and 90°; S13: Measure the curvature of the pile surface using a laser rangefinder and adjust the bracket angle so that the axis of the sensor unit group always matches the normal direction of the pile surface within a predetermined range.

3. The method for real-time detection of local scour of a wind turbine monopile based on multi-frequency acoustic tomography according to claim 1, characterized in that, S2 specifically includes: S21: Let the center frequency of the low-frequency penetrating wave be... Its frequency range is Let the center frequency of the high-frequency surface wave be... Its frequency range is ;in ; S22: Within a set single transmission cycle T, low-frequency penetrating waves are transmitted during the first 0.3T period; high-frequency surface waves are transmitted during the last 0.7T period. S23: Through the phased array of the transmitting unit group, low-frequency penetrating waves are focused on the mud layer infiltration zone at a depth of 0-5m around the pile, while high-frequency surface waves cover the seabed surface area within a range of 0-2m around the pile.

4. The method for real-time detection of local scour of a wind turbine monopile based on multi-frequency acoustic tomography according to claim 1, characterized in that, S3 specifically includes: S31: The scattered wave signal reflected from the pile-seabed interface is acquired by the receiving unit group, and the scattered wave signal includes reflection components from low-frequency penetrating waves and high-frequency surface waves; S32: The time-frequency separation algorithm is used to process the acquired scattered wave signal. Specifically, the scattered wave signal is first decomposed according to the characteristics of the time domain and frequency domain; then, the low-frequency scattered wave and high-frequency scattered wave corresponding to the S2 frequency are identified and extracted.

5. A method for real-time detection of local scour of a wind turbine monopile based on multi-frequency acoustic tomography according to claim 4, characterized in that, The scattered wave signal is decomposed according to its characteristics in the time and frequency domains, including: S321: Based on the short-time Fourier transform, the collected scattered wave signal is divided into time windows, and the Fourier transform is performed on the signal in each time window to obtain the frequency components in the corresponding time period. S322: Based on the Fourier transform, the time-domain signal is converted into a frequency-domain signal; by integrating the time-domain signal, the signal components in the frequency domain are obtained.

6. The method for real-time detection of local scour of a wind turbine monopile based on multi-frequency acoustic tomography according to claim 4, characterized in that, The process of identifying and extracting the low-frequency and high-frequency scattered waves corresponding to the S2 frequency includes: Signal frequency range division: Based on the frequency range of the low-frequency and high-frequency signals generated in S2, the frequency intervals of the low-frequency and high-frequency signals are set; Frequency filtering: Using low-pass and high-pass filters to process frequency domain signals. Filtering is performed to separate the low-frequency components. and high-frequency components ; Signal reconstruction: Reconstructing the filtered low-frequency components and high-frequency components The inverse Fourier transform is performed back to the time domain signal to obtain the corresponding low-frequency and high-frequency time domain signals, and finally the low-frequency and high-frequency scattered wave signals are obtained.

7. The method for real-time detection of local scour of a wind turbine monopile based on multi-frequency acoustic tomography according to claim 1, characterized in that, The construction of the acoustic tomography matrix for mud penetration includes: Sound wave path calculation: Based on the time delay distribution data of low-frequency scattered waves, combined with the sound velocity information of seawater and silt, the propagation path of sound waves from the emission point to the receiving point is calculated by the curved ray tracing inversion algorithm, and the specific coordinates of each path are determined. Path correction: Based on the measured flow velocity and seawater viscosity, the radius of curvature of the sound wave path is adjusted using the Reynolds number to correct the propagation characteristics of the path and compensate for the influence of fluid dynamics on sound wave propagation. Path Discretization and Attenuation Rate Calculation: The corrected acoustic path is discretized into a voxel grid, and the energy attenuation rate of each path is calculated. A mud-penetration acoustic tomography matrix is ​​generated using the ratio of the transmitted and received wave amplitudes. .

8. The method for real-time detection of local scour of a wind turbine monopile based on multi-frequency acoustic tomography according to claim 1, characterized in that, The construction of the surface topography acoustic tomography matrix based on high-frequency scattered waves includes: Finite-difference time-domain beamforming: The wave field intensity of each three-dimensional grid point is calculated by using the finite-difference time-domain method. The wave field intensity of the grid point is obtained by multiplying the signal of each receiving unit group by its weight after the propagation time correction, and then summing the contributions of all receiving units by weight. Wave field intensity extraction and normalization: Extract wave field intensity data of the seabed surface layer and normalize it to the [0,1] interval; Generate the acoustic tomography matrix: Construct the surface topography acoustic tomography matrix using normalized wave field intensity data. .

9. A real-time detection method for local scour of a wind turbine monopile based on multi-frequency acoustic tomography according to claim 8, characterized in that, S5 specifically includes: S51: Obtain the acoustic tomography matrix of mud penetration constructed in S4 and surface morphology acoustic tomography matrix This ensures that the two are aligned in spatial coordinates; S52: Yes and The elements in the array are standardized to ensure that their numerical ranges are consistent. S53: Using a weighted average method, the standardized... and The data are fused, and the resulting 3D image of the scour state is calculated. The expression is: ,in, and These are the weighting coefficients for the mud layer and the seabed surface layer, respectively. The generated 3D image of the scouring state; S54: Merge the matrix Converted into a three-dimensional scour state image.

10. A real-time detection method for local scour of a wind turbine monopile based on multi-frequency acoustic tomography according to claim 1, characterized in that, S6 specifically includes: S61: Extract acoustic wave propagation delay data from the 3D image of the scour state generated by S5, and identify the regions where the acoustic wave propagation delay changes significantly by setting a delay abrupt change threshold, i.e., delay abrupt change regions. S62: After identifying the time delay abrupt change region, locate its spatial coordinates, determine the boundary of the abrupt change region, and compare it with known seabed morphology data to locate the specific scour pit location. S63: Extract the outline of the abrupt change region to determine the shape and size of the scour pit.

Images