Three-dimensional sonar offshore pile foundation online monitoring method and system

Through the three-dimensional sonar online monitoring system, combined with the Internet of Things and AI technology, the problem that traditional monitoring systems cannot respond to extreme weather and tide changes in a timely manner, real-time monitoring and trend evaluation of local erosion of bridges or pile foundations has been achieved, and the The timeliness and accuracy of monitoring ensures the safety of bridges or bearings.

CN117908033BActive Publication Date: 2024-11-15长三角国创超声(上海)有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410085618.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-01-22
Publication Date
2024-11-15
Estimated Expiration
2044-01-22

AI Technical Summary

Technical Problem

The monitoring systems of traditional bridges or bearings rely on manual regular inspections and are unable to respond to extreme weather and tide changes in time, resulting in a single, untimely and untraceable monitoring data, making it difficult to effectively evaluate the overall trend of local erosion of pile foundations.

Method used

The online monitoring method of offshore pile foundation of three-dimensional sonar is used, combined with the Internet of Things and artificial intelligence technology, and the sound wave signal is transmitted to the seabed through 3D sonar, the target position is estimated using the joint estimation algorithm of TOA and DOA, the depth data is calculated in combination with the trigonometric function, and the edge calculation is carried out. Conduct data compression, analysis and early warning with AI to achieve multi-dimensional comparison and trend analysis.

Benefits of technology

Real-time monitoring and trend evaluation of local erosion of bridge or platform pile foundations has been achieved, which improves the timeliness, traceability and accuracy of monitoring, enhances the ability to respond to extreme weather and sea tides, and ensures the safety of bridge or platform. Sex and reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117908033B_ABST
    Figure CN117908033B_ABST
Patent Text Reader

Abstract

The invention discloses a three-dimensional sonar offshore pile foundation online monitoring method and system, comprising: emitting a highly directional sound wave signal to the seabed / river bottom based on a 3D sonar; when the obstacle encountered is the seabed / river bottom, reflecting the echo signal and obtaining a detection threshold of the seabed signal, obtaining the relative distance between the seabed / river bottom and the 3D sonar based on the detection threshold, and obtaining distance data from the sound wave emission point to the sound wave arrival end point; when the distance data is obtained, simultaneously obtaining the arrival time and arrival direction of the current echo signal and obtaining the beam pitch angle and azimuth angle corresponding to the current echo signal; estimating the target position parameters of the echo based on the obtained arrival time and arrival direction of the current echo signal and based on a joint estimation algorithm of TOA and DOA; and calculating the target depth data by trigonometric function in combination with the target position parameters to improve monitoring efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of underwater bridge detection, and in particular to a three-dimensional sonar offshore pile foundation online monitoring method and system. Background Art

[0002] The local scouring of bridge piers or abutments is one of the main reasons for water damage to bridges or abutments. The scouring depth is an important basis for determining the burial depth of the bridge piers or abutments, which is directly related to the safety of the bridge or abutments. Therefore, timely understanding and mastering the scouring status of each bridge pier or abutment is of great significance for bridge design, construction, flood control and safe navigation.

[0003] Traditional monitoring systems only rely on manual monitoring once a year or every few years. They lack the timeliness, cognition, and traceability of the monitored objects in the face of extreme weather and tides. The data from each three-dimensional sonar monitoring point is not timely, and the data obtained is single.

[0004] Therefore, it is necessary to propose a three-dimensional sonar online monitoring method for offshore pile foundations to improve the monitoring efficiency. Summary of the invention

[0005] The purpose of the present invention is to provide a three-dimensional sonar online monitoring method and system for offshore pile foundations. By combining with the Internet of Things technology and artificial intelligence technology, a three-dimensional sonar online monitoring system is formed, so that the data of each three-dimensional sonar monitoring point are uniformly gathered on the online monitoring system platform, breaking through the limitation of single data, and realizing multi-dimensional comparative analysis and trend analysis of the monitoring data of local scour of pile foundations, providing a system platform support for the overall trend evaluation of local scour of pile foundations.

[0006] The present invention also provides a three-dimensional sonar offshore pile foundation online monitoring method, which is used in a three-dimensional sonar offshore pile foundation online monitoring system, comprising:

[0007] Based on 3D sonar, it emits highly directional sound wave signals to the seabed / river bottom;

[0008] Determine whether the acoustic wave signal encounters an obstacle, and when the obstacle encountered is the seabed / river bottom, reflect the echo signal and obtain a detection threshold of the seabed signal, obtain the relative distance between the seabed / river bottom and the 3D sonar based on the detection threshold, and obtain the distance data from the acoustic wave emission point to the acoustic wave arrival end point;

[0009] When the distance data is obtained, the arrival time and arrival direction of the current echo signal are simultaneously obtained to obtain the beam pitch angle and azimuth angle corresponding to the current echo signal;

[0010] According to the acquired arrival time and arrival direction of the current echo signal, the target position parameters of the echo are estimated based on the joint estimation algorithm of TOA and DOA;

[0011] The target depth data is calculated by trigonometric function in combination with the target position parameters.

[0012] Preferably, the 3D sonar comprises:

[0013] A stepper motor controller is used to subdivide the drive according to the azimuth angle control requirements and the pitch angle control requirements, and adjust the rotation speed of the motor to achieve precise control of the azimuth angle and pitch angle;

[0014] The transducer is connected to the stepper motor controller through a transmission shaft to form a 3D pan-tilt structure, and is used to control the stepper motor controller to drive the corresponding motor to rotate around the transmission shaft according to a preset azimuth angle range, drive the transducer to rotate for 360° scanning, and feedback the azimuth angle information through the sensor to complete the azimuth angle measurement. At the same time, according to a preset pitch angle, the stepper motor controller is controlled to drive the corresponding motor to rotate around the transmission shaft, drive the transducer to rotate for 90° scanning, and feedback the pitch angle information through the sensor to complete the pitch angle measurement;

[0015] Wherein, each subdivision angle of the stepper motor controller can be accurate to 0.225°.

[0016] Preferably, the internal circuit structure of the 3D sonar includes a signal processor, a transmitting drive circuit, a data acquisition circuit, a power amplifier, a signal conditioning circuit, a transceiver conversion switch and a transducer. The signal processor controls the transmitting signal to be emitted from the transmitting drive circuit and the power amplifier, and the transmitting signal is transmitted to the transducer through the transceiver conversion switch. The transducer converts the electrical signal corresponding to the transmitting signal into an acoustic wave signal and emits it to the seabed / river bottom. The acoustic wave signal reflects an echo signal during propagation in the water. After being received by the transducer, the echo signal is transmitted to the signal processor through the transceiver conversion switch, the signal conditioner and the data collector for reading, analysis and processing, and finally generates the depth information of the target location.

[0017] When the machine is turned on, under the control of the signal processor, the azimuth and elevation angles are reset to zero;

[0018] Start sonar scanning according to the set working parameters, and divide the seabed / river bottom sonar coverage area into multiple grid areas;

[0019] The center of each grid area is mapped to the azimuth and pitch angle of the spherical coordinate system, and the corresponding motor is controlled by the stepper motor controller to align the sonar beam with the center of the grid area, and the corresponding scanning and data acquisition operations are performed on the sonar beam.

[0020] Preferably, the TOA and DOA-based joint estimation algorithm estimates the target position parameters of the echo, including:

[0021] The joint weighted time average and azimuth deviation indication are used to obtain the precise arrival time of the echo signal and calculate the depth data at the current incident angle. The specific steps are as follows:

[0022] When the incident angle is less than the preset angle threshold, the weighted time averaging algorithm is used to obtain the amplitude sequence corresponding to the echo signal, and the sampling points exceeding the threshold are selected from the received echo amplitude sequence, and the TOA value of the current echo signal is obtained by calculating the time-weighted average of the amplitude of the sampling points;

[0023] When the incident angle of the echo is greater than or equal to the set angle threshold, the BDI algorithm is used to obtain time slice samples. The received echo signals are arranged into a series of spatial grids according to the beam angle direction. For each time slice, the grids whose energy exceeds the threshold are selected to make an initial judgment on the number of echoes. The energy of a group of grids suspected of echoes is fitted using a second-order polynomial to obtain the DOA value of the current echo signal.

[0024] Preferably, the TOA and DOA-based joint estimation algorithm estimates the target position parameters of the echo further includes:

[0025] When the angle of arrival of the echo signal is θ, the beam angle and the front and rear beam angles are: β, β l and β r , and satisfy β-β l =β r -β=Δβ;

[0026] The signal form of the three beams is expressed as:

[0027] The echo amplitude of the previous beam angle is

[0028] The echo amplitude at the current beam angle is

[0029] The echo amplitude of the latter beam angle is

[0030] Among them: a l , a, a r are the echo amplitudes of the previous beam, the current beam and the next beam respectively, and k is a constant;

[0031] Subtract the logarithms of the echo amplitude of the previous beam angle, the echo amplitude of the current beam angle, and the echo amplitude of the next beam angle, and then sort them out to get the angle formula of the echo signal arrival as follows:

[0032]

[0033] Where: Δa r =ln(a r )-ln(a), Δa l =ln(a)-ln(a l ),

[0034] When the echo angle is known, the echo arrival angle DOA is calculated according to the angle formula of the echo signal arrival.

[0035] Preferably, calculating the target depth data by trigonometric function in combination with the target position parameters comprises:

[0036] Establish a three-dimensional coordinate system, with the sonar probe as the coordinate origin O(0, 0, 0), confirm the direction of the azimuth angle α and the pitch angle β of the transducer to transmit sound waves, and receive and detect echoes in this direction;

[0037] When the sound wave encounters the target T (x0, y0, z0) and is scattered, the distance R between OT is calculated;

[0038] The projection (x1, y1, z1) of the target T in the coordinate system is calculated according to the trigonometric function. The formula is as follows:

[0039] x1=R*sin(β)*cos(α);

[0040] y1=R*sin(β)*sin(α);

[0041] z1=R*cos(β);

[0042] Among them, (x1, y1) is the position / projection of target T on the horizontal plane, and z1 is the depth of target T.

[0043] Preferably, the transducer includes a plurality of independent circular rings cut from the inside to the outside, and the areas of the plurality of distributed circular rings are obtained by cosine weighting, so that the energy applied to the plurality of circular rings is distributed according to the cosine weighting coefficient, thereby achieving the reduction of side lobes, wherein the cosine weighting formula is as follows: W(n)=1 / 2(1+cos(2πn / (N-1))), N is 16, n is 0, 1, 2, 3, 4, and the corresponding area ratios of S1 to S5 are as follows: S1:S2:S3:S4:S5=1:0.96:0.85:0.69:0.5.

[0044] The present invention also provides a three-dimensional sonar offshore pile foundation online monitoring system, comprising:

[0045] The acquisition module includes a 3D sonar side-scan unit for local scouring of each pile foundation, which is used to perform vector calculations on the acquired side-scan data and compress the monitoring data capacity through an efficient compression algorithm for 3D sonar data based on edge computing;

[0046] The transmission module is used to combine the actual communication conditions on site and support wired network or wireless communication to achieve data transmission and analysis;

[0047] A storage module is used to receive three-dimensional sonar data and extract three-dimensional sonar feature compression data for vector reconstruction, and automatically generate and store corresponding three-dimensional sonar monitoring dot cloud images;

[0048] The service module includes a data cache server, a data docking server, an AI artificial intelligence server and a web application server. The data cache server is used to further optimize the vector structure data.

[0049] Data docking server: docking with other system data to obtain more dimensional information and provide data sources of various dimensions for data mining;

[0050] AI artificial intelligence server: The vector data generated by each 3D sonar side scan monitoring is quantified through the feature model to form regression vectors, low-level height vectors, and high-level height vectors. The corresponding vectors are selected as training samples for the target elements and input into the machine learning algorithm model for training to obtain quantitative, qualitative and trend analysis data related to pile foundation scour depth, scour type, morphology and correlation factors;

[0051] Web application server: used to provide service support for visual data display;

[0052] The application module is used to realize the aggregation, parsing, storage, analysis and sonar image display of local scour data of pile foundation through interface operation and display.

[0053] Preferably, the function implementation steps of the AI ​​artificial intelligence server include:

[0054] Acquire pile foundation scour depth data, wherein the pile foundation scour depth data includes low-level height vector data and high-level height vector data;

[0055] Based on the edge computing method, the low-level height vector and the high-level height vector are distinguished, and the pile foundation scouring depth data is compared with the preset depth threshold. If the current pile foundation scouring depth data is greater than the preset depth threshold, it is determined to be high-level height vector data; if the current pile foundation scouring depth data is less than or equal to the preset depth threshold, it is determined to be low-level height vector data, and the high-level height vector data is uploaded to the data queue in the order of priority, and the high-level height vector data is processed first and then the low-level height vector data;

[0056] The first pile foundation scouring depth curve is obtained by using a linear regression algorithm with the pile foundation scouring relative depth and time T as factors. When affected by external factors, the second pile foundation scouring relative depth data uploaded by the three-dimensional sonar is greater than the first pile foundation scouring depth data obtained by the first pile foundation scouring depth curve. At this time, it is determined to be an exogenous mutation curve, and the external factors at least include typhoons.

[0057] Each pile foundation is equipped with an attention counter. When a high-height vector appears, the corresponding attention counter value is increased to obtain the attention count data of each pile foundation. The attention count data of each pile foundation is classified and aggregated through the K-means algorithm to obtain the number of pile foundations with the highest attention.

[0058] If the proportion of high-attention pile foundations to the total number of pile foundations of the bridge is greater than the preset attention threshold, it is judged that the bridge safety attention is relatively high.

[0059] The present invention provides an electronic device, characterized in that it includes:

[0060] A memory, the memory being used to store a processing program;

[0061] A processor, wherein when executing the processing program, the processor implements the three-dimensional sonar offshore pile foundation online monitoring method as described in the embodiment of the present invention.

[0062] With respect to the prior art, the present invention has the following beneficial effects:

[0063] The sonar online monitoring method and system for cross-sea bridges or offshore wind turbine piers of the present invention installs one or more sonars on piers or piers to obtain local scour depth data of each pier or pier through real-time scanning and monitoring, and transmits the data of each sonar back to a data monitoring center through the Internet of Things technology for data display, analysis and early warning, thereby timely controlling the local scour depth and change trend of each pier or pier of the cross-sea bridge or offshore wind turbine pier, overturning the traditional mode of relying solely on manual monitoring once a year or several years, greatly improving the timeliness, cognition and traceability of the monitored object in the face of extreme weather and tides, thereby greatly improving the ability of the cross-sea bridge or offshore wind turbine pier to resist extreme weather and tides, ensuring its reliability and safety. The sonar online monitoring system for cross-sea bridges or offshore wind turbine piers is innovative, leading and exemplary. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a schematic diagram of the process of the three-dimensional sonar offshore pile foundation online monitoring method described in one embodiment of the present invention;

[0065] Figure 2 This is a control block diagram of a 3D sonar pan / tilt platform in one embodiment of the present invention;

[0066] Figure 3 This is a structural diagram of a 3D sonar pan-tilt platform in one embodiment of the present invention;

[0067] Figure 4 This is a control block diagram of a 3D sonar pan / tilt platform in one embodiment of the present invention;

[0068] Figure 5 This is a schematic diagram of pile foundation seabed sonar detection in one embodiment of the present invention;

[0069] Figure 6 This is an example diagram of echo detection in one embodiment of the present invention;

[0070] Figure 7 This is an example diagram of depth data calculation in one embodiment of the present invention;

[0071] Figure 8 This is an example diagram of a transducer cutting process similar to cosine weighting in one embodiment of the present invention;

[0072] Fig. 9 It is a schematic diagram of a three-dimensional sonar offshore pile foundation online monitoring system according to an embodiment of the present invention;

[0073] Fig.10 Schematic diagram of the steps of a method for predicting the impact of typhoons and earthquakes on pile foundation depth based on a fully connected neural network in one embodiment of the present invention. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0076] It should be noted that the concepts such as "once" and "second" mentioned in the present application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0077] It should be noted that the modifications of "one" and "plurality" mentioned in the disclosure of this application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0078] Embodiment 1

[0079] The present invention provides a three-dimensional sonar offshore pile foundation online monitoring method, which is implemented based on the following three-dimensional sonar offshore pile foundation online monitoring system. The physical device of the system integrates sonar detection, data imaging, Internet of Things, various forms of power supply, communication, data analysis and mining artificial intelligence, information technology and other aspects, including: monitoring system, wireless modem, local power supply, solar power supply, sonar data collector, cable, 3D sonar, and the specific monitoring steps include:

[0080] like Figure 1 As shown, based on 3D sonar, a highly directional sound wave signal is emitted to the seabed / river bottom;

[0081] Determine whether the acoustic wave signal encounters an obstacle, and when the obstacle encountered is the seabed / river bottom, reflect the echo signal and obtain a detection threshold of the seabed signal, obtain the relative distance between the seabed / river bottom and the 3D sonar based on the detection threshold, and obtain the distance data from the acoustic wave emission point to the acoustic wave arrival end point;

[0082] When the distance data is obtained, the arrival time and arrival direction of the current echo signal are simultaneously obtained to obtain the beam pitch angle and azimuth angle corresponding to the current echo signal;

[0083] According to the acquired arrival time and arrival direction of the current echo signal, the target position parameters of the echo are estimated based on the joint estimation algorithm of TOA and DOA;

[0084] The target depth data is calculated by trigonometric function in combination with the target position parameters.

[0085] In one embodiment, see Figure 2 As shown, the 3D sonar comprises:

[0086] The stepper motor controller 201 is used to subdivide the drive according to the azimuth angle control requirements and the pitch angle control requirements, and adjust the rotation speed of the motor to achieve precise control of the azimuth angle and the pitch angle;

[0087] The transducer 202 is connected to the stepper motor controller 201 through a transmission shaft to form a 3D gimbal structure, which is used to control the stepper motor controller 201 to drive the corresponding motor to rotate around the transmission shaft 203 according to a preset azimuth range, drive the transducer 202 to rotate for 360° scanning, and feedback the azimuth information through the sensor to complete the azimuth measurement. At the same time, according to the preset pitch angle, the stepper motor controller 201 is controlled to drive the corresponding motor to rotate around the transmission shaft 203, drive the transducer 202 to rotate for 90° scanning, and feedback the pitch angle information through the sensor to complete the pitch angle measurement; wherein, the dual-axis combination realizes high-precision scanning of the half space, and each subdivision angle of the stepper motor controller 201 can be accurate to 0.225°. It can be understood that in order to detect the 180° half space below the water surface, the azimuth angle can be understood as the transmitting beam rotating one circle in the horizontal direction, that is, after 360°, the pitch angle is stepped by one angle again, and then rotated 360° until the half space below the water surface is completely scanned on one side, see Figure 3 shown.

[0088] like Figure 4As shown, the internal circuit structure of the 3D sonar includes a signal processor, a transmitting drive circuit, a data acquisition circuit, a power amplifier, a signal conditioning circuit, a transceiver conversion switch and a transducer. The signal processor controls the transmitting signal to be emitted from the transmitting drive circuit and the power amplifier, and the transmitting signal is transmitted to the transducer through the transceiver conversion switch. The transducer converts the electrical signal corresponding to the transmitting signal into an acoustic wave signal and emits it to the seabed / river bottom. The acoustic wave signal reflects an echo signal during propagation in the water. After being received by the transducer, the echo signal is transmitted to the signal processor through the transceiver conversion switch, the signal conditioner and the data collector for reading, analysis and processing, and finally generates the depth information of the target location.

[0089] When the machine is turned on, under the control of the signal processor, the azimuth and elevation angles are reset to zero;

[0090] Start sonar scanning according to the set working parameters, and divide the seabed / river bottom sonar coverage area into multiple grid areas;

[0091] The center of each grid area is mapped to the azimuth and pitch angle of the spherical coordinate system, and the corresponding motor is controlled by the stepper motor controller to align the sonar beam with the center of the grid area, and the corresponding scanning and data acquisition operations are performed on the sonar beam.

[0092] Those skilled in the art will appreciate that the circuit diagram of the 3D sonar is as follows: Figure 4 As shown. The signal processor S401 is the core of the circuit, which contains control software, arithmetic processing unit, data storage unit and signal transmission unit. When the machine is turned on, under the control of the signal processor software, the 3D sonar performs azimuth and pitch angle zeroing, and then starts sonar scanning according to the set working parameters. The seabed / river bottom sonar coverage area is divided into rectangular grid blocks to facilitate the post-processing of sonar data. The side length of the square is related to the current depth, but the maximum does not exceed 1m. The center of each square is mapped to the azimuth and pitch angle of the spherical coordinate system, and the stepper motor controller controls the motor to align the sonar beam with the center of the square. The signal processor S401 controls the transmitting driver S404 and the power amplifier S405 to transmit sound waves. The transmitting signal is applied to the transducer S407 through the transceiver conversion switch S406, and then the transducer converts the electrical signal into an acoustic signal and transmits it. When the acoustic signal propagates in the water, it will reflect the sound waves when it encounters an object. After being received by the transducer, the sound waves pass through the transceiver conversion switch S406, the signal conditioner S403, and the data collector S402, and then the signal processor reads, analyzes and processes them, and finally generates the depth information of the square. Data processing uses TOA estimation and DOA estimation algorithms.

[0093] like Figure 5As shown in the figure, assuming that the 3D sonar transducer emits a pulse in one direction, such as an azimuth of 0° and an elevation angle of β, the sound wave pulse returns after encountering the seabed, is converted into an electrical signal by the 3D sonar transducer, and then amplified and output by the receiver. Since the beam emitted by the transducer has a width, the echo signal that first reaches the seabed is not the echo signal at the center of the beam we want, but the echo signal at the edge of the beam. The receiver outputs a pulse signal with a certain width. The signal width is a function of the elevation angle β. The larger β is, the wider the width is.

[0094] In the case of a small pitch angle, the sound wave is basically incident vertically on the seabed. In this case, the center of the beam is the first to reach the seabed. At this time, there is no need to estimate the azimuth of the echo arrival, only the TOA (Time of Arrival) of the echo arrival needs to be estimated.

[0095] In the case of large pitch angles, the edge of the sound wave beam will reach the seabed first. At this time, the echo angle is not the echo angle we need. In order to accurately obtain the true echo angle and the true arrival time of the signal, it is necessary to estimate it through signal processing technology. We need to adopt the method of joint estimation of DOA (Direction of Arrival) and TOA.

[0096] Regardless of which method is used, we will eventually obtain the precise pitch angle β and the precise echo arrival time t. Based on the propagation speed c of sound waves in water, we can obtain the slant distance R=t*c / 2 between the 3D sonar location and the point where the beam meets the seabed. This allows us to calculate the height H=R*cos(β) between the horizontal plane of the 3D sonar location and the point where the beam meets the seabed.

[0097] By controlling the rotation of the pitch control motor, the pitch angle can be rotated from 0° to 90°. Each rotation of 0.9° can obtain a height information, and then the azimuth motor is controlled to rotate 0.9°, and the pitch angle is repeatedly rotated from 0° to 90° until the azimuth rotates 360°, thereby obtaining the height data of the underwater half space. The 3D sonar sends the obtained height data to the host computer, which forms a point cloud map or performs data processing.

[0098] Due to the irregularity of the sea surface and the seabed, the signal is scattered at the reflection point. Therefore, the reflection point is regarded as a distributed signal source, and the central wave direction of the scattered signal is used as DOA, and the average wave arrival time is used as TOA. Finally, the arrival time and arrival angle are used to estimate the position parameters of the target. The simulation results show that this method is superior to the traditional straight-line transmission model in terms of positioning accuracy and robustness. TOA refers to time of arrival, which refers to the time estimate of the arrival time of the echo, and DOA is direction of arrival, which refers to the direction reached by the echo. In this embodiment, only by measuring the arrival time of the echo and the arrival direction of the echo can the location information of the echo be calculated, thereby calculating the depth of the position. TOA estimation and DOA estimation algorithm: The joint weighted mean time (Weighted Mean Time, referred to as WMT) and bearing deviation indicator (Bearing Deviation Indicator, referred to as BDI) are used to obtain the accurate echo arrival time and calculate the depth data under the current incident angle. The specific implementation steps include:

[0099] The joint weighted time average and azimuth deviation indication are used to obtain the precise arrival time of the echo signal and calculate the depth data at the current incident angle. The specific steps are as follows:

[0100] When the incident angle is less than 15°, the weighted time average algorithm is used to obtain the amplitude sequence corresponding to the echo signal, and the sampling points exceeding the threshold are selected from the received echo amplitude sequence, and the TOA value of the current echo signal is obtained by calculating the time weighted average of the amplitude of the sampling points;

[0101] When the incident angle of the echo is greater than or equal to 15°, the BDI algorithm is used to obtain time slice samples. The received echo signals are arranged into a series of spatial grids according to the beam angle direction. For each time slice, the grids whose energy exceeds the threshold are selected to make an initial judgment on the number of echoes. The energy of a group of suspected echo grids is fitted using a second-order polynomial to obtain the DOA value of the current echo signal. See Figure 6 shown.

[0102] In this embodiment, the 3D sonar uses a motor to control the pointing angle of the transducer, so assuming that the beam angle is known, the purpose of echo DOA correction can be achieved by comparing the amplitudes of two adjacent echoes.

[0103] Assuming that the angle of arrival of the echo signal is θ, the beam angle and the front and rear beam angles are: β, β l and β r , and β-β l =β r -β=Δβ,

[0104] The signal form of the three beams can be expressed as:

[0105]

[0106]

[0107]

[0108] Among them: a l , a, a r The echo amplitude of the previous beam, the current beam and the next beam, k is a constant.

[0109] Subtract the logarithms of equations (1), (2), and (3) and sort them out to get:

[0110]

[0111] Where: Δa r =ln(a r )-ln(a), Δa l =ln(a)-ln(a l )

[0112] When the echo angle is known, the echo arrival angle DOA can be calculated according to formula (4).

[0113] like Figure 7 As shown, the depth data calculation in this embodiment is described as follows:

[0114] Calculating target depth data by trigonometric function in combination with the target position parameters includes:

[0115] Establish a three-dimensional coordinate system, with the sonar probe as the coordinate origin O(0,0,0), confirm the direction of the transducer's azimuth angle α and elevation angle β to transmit sound waves, and receive and detect echoes in this direction;

[0116] When the sound wave encounters the target T (x0, y0, z0) and is scattered, the distance R between OT is calculated;

[0117] The projection (x1, y1, z1) of the target T in the coordinate system is calculated according to the trigonometric function. The formula is as follows:

[0118] x1=R*sin(β)*cos(α);

[0119] y1=R*sin(β)*sin(α);

[0120] z1=R*cos(β);

[0121] Among them, (x1, y1) is the position / projection of target T on the horizontal plane, and z1 is the depth of target T.

[0122] In various embodiments of the present application, the sonar sound-transmitting layer of the 3D sonar is provided with a protective layer to prevent the attachment and growth of marine organisms, so as to meet the needs of long-term online detection of the 3D sonar. This embodiment adopts a broadband transducer (S204) made of composite materials for electroacoustic energy conversion. A cutting process similar to cosine weighting is adopted, such as Figure 8 As shown in FIG. 1 , the relationship between the ceramic sheet areas S1, S2, S3, S4 and S5 obeys the cosine law. This transducer cutting process can obtain very low transmitting and receiving side lobes, which can effectively suppress the energy received by the side lobes, thereby forming a very clear acoustic image.

[0123] The directivity of the disk transducer of 3D sonar determines the accuracy of measurement. The ideal directivity of the transducer should have a very small beam width and no side lobes. But in fact, due to the limitation of transducer size, the directivity of the transducer is related to frequency and size. The directivity formula is as follows:

[0124]

[0125] Where: J1: first-order Bessel function, D: transducer diameter, λ: wavelength.

[0126] It can be seen that the transducer directivity has side lobes. When working at a large pitch angle, the side lobes may reach the seabed before the main lobe, causing sonar misjudgment and measurement errors. The disc transducer is cut into multiple independent rings, and the ring area is obtained by cosine weighting, so that the energy applied to these rings is distributed according to the cosine weighting coefficient, thereby achieving the purpose of reducing the side lobes. The cosine weighting formula is as follows:

[0127]

[0128] Among them: N is 16, n is 0, 1, 2, 3, 4, and the corresponding area ratios of S1 to S5 are as follows: S1: S2: S3: S4: S5 = 1: 0.96: 0.85: 0.69: 0.5.

[0129] The transducer used in this embodiment can be a transceiver combined with a transmitter, which can further improve the angular resolution and further reduce the side lobe. For online application scenarios such as pile foundations, high-precision seabed topography data can be obtained through slow high-precision scanning technology, reducing the complexity of the system and improving the economy of the system.

[0130] The application of parametric array technology can obtain high-definition seabed / riverbed topography images while acquiring layered structural information of seabed geology, providing data support for scour analysis around pile foundations.

[0131] Embodiment 2

[0132] Based on the same inventive concept, the present invention also provides a three-dimensional sonar offshore pile foundation online monitoring system. The three-dimensional sonar is combined with the Internet of Things technology and artificial intelligence technology to form a three-dimensional sonar online monitoring system, so that the data of each three-dimensional sonar monitoring point is unified on the online monitoring system platform, breaking through the limitation of single data, and realizing multi-dimensional comparative analysis and trend analysis of the monitoring data of local scour of the pile foundation, providing a system platform support for the overall trend assessment of local scour of the pile foundation.

[0133] The architecture of the three-dimensional sonar online monitoring system based on the Internet of Things technology is as follows: Fig. 9 The entire online monitoring system is divided into five layers, namely, the acquisition layer, transmission layer, storage layer, service layer, and application layer, namely, the functions implemented by the following acquisition module, transmission module, storage module, service module, and application module. The specific description is as follows:

[0134] The acquisition module includes a 3D sonar side-scan unit for local scouring of each pile foundation, which is used to perform vector calculations on the acquired side-scan data. The monitoring data capacity is greatly compressed by using an efficient compression algorithm for 3D sonar data based on edge computing, which reduces the pressure on the transmission layer and improves data transmission efficiency.

[0135] The transmission module is used to combine the actual communication conditions on site and support wired network or wireless communication methods to achieve data transmission and analysis, such as wireless 4G / LoRa / Beidou / satellite, etc.;

[0136] The storage module is used to receive 3D sonar data and extract 3D sonar feature compressed data for vector reconstruction, and automatically generate and store the corresponding 3D sonar monitoring dot cloud map to facilitate data analysis, mining and system access at the service layer;

[0137] The service module includes a data cache server, a data docking server, an AI artificial intelligence server and a web application server. The data cache server is used to perform targeted algorithm optimization on the vector structure data cache server, which will accelerate the calculation speed of algorithms such as vector data aggregation and classification; the data docking server: docks with other system data to obtain more dimensional information and provide data sources of various dimensions for data mining; the AI ​​artificial intelligence server: quantifies the vector data formed by each three-dimensional sonar side-scan monitoring through a feature model to form regression vectors, low-level height vectors, and high-level height vectors, and selects the corresponding vectors for target elements as training samples to input into the machine learning algorithm model for training, and obtains quantitative, qualitative and trend analysis data related to pile foundation scouring depth, scouring type, morphology, and correlation factors; the target elements include relevant For height, number of occurrences, density and volatility, machine learning algorithm models such as linear regression, K-means, decision tree, random forest, etc. are used; feature classification forms a unique three-dimensional sonar absolute vector and relative vector index tree, which accelerates the data loading efficiency of the data cache server; logistic regression, decision tree and other algorithms form a pile foundation scour depth classification warning model, and early classification warning is provided for risky pile foundations through the online monitoring system; web application server: used to provide service support for visual data display; application module, used to realize the aggregation, parsing, storage, analysis and sonar image display of local scour data of pile foundation through interface operation and display. AI artificial intelligence technology provides technical support for timely discovery of pile foundation safety hazards and automatic warning through models and algorithms, making up for the timely control of local scour hazards of pile foundation and assessment blind spots, and further improving the safety of pile foundation.

[0138] Specifically, the steps for implementing the functions of the AI ​​artificial intelligence server include:

[0139] Acquire pile foundation scour depth data, wherein the pile foundation scour depth data includes low-level height vector data and high-level height vector data;

[0140] Based on the edge computing method, the low-level height vector and the high-level height vector are distinguished, and the pile foundation scouring depth data is compared with the preset depth threshold. If the current pile foundation scouring depth data is greater than the preset depth threshold, it is determined to be high-level height vector data; if the current pile foundation scouring depth data is less than or equal to the preset depth threshold, it is determined to be low-level height vector data, and the high-level height vector data is uploaded to the data queue in the order of priority, and the high-level height vector data is processed first and then the low-level height vector data;

[0141] The first pile foundation scour depth curve is obtained by a linear regression algorithm using the relative depth of pile foundation scour and time T as factors. When affected by external factors, the second pile foundation scour relative depth data uploaded by three-dimensional sonar is greater than the first pile foundation scour depth data obtained by the first pile foundation scour depth curve. At this time, it is determined to be an exogenous mutation curve, and the external factors at least include typhoons; the exogenous mutation curve can be a typhoon mutation curve.

[0142] Each pile foundation is equipped with an attention counter. When a high-height vector appears, the corresponding attention counter value is increased to obtain the attention count data of each pile foundation. The attention count data of each pile foundation is classified and aggregated through the K-means algorithm to obtain the number of pile foundations with the highest attention.

[0143] If the proportion of high-concern pile foundations to the total number of pile foundations of the bridge is greater than the preset concern threshold, it is determined that the safety concern of the bridge is relatively high, wherein each bridge includes multiple pile foundations.

[0144] Those skilled in the art can understand that this embodiment uses edge computing technology in the three-dimensional sonar controller to distinguish low-level height vectors and high-level height vectors. High-level height vector data is defined as data with a relatively deep single scour depth, and high-level height vector data is uploaded preferentially in the upload data queue, so that the artificial intelligence server can obtain emergency data processing preferentially. Generally, pile foundation scour is a slow process, and the change in the depth of two adjacent scours of the same pile foundation is generally very small. The relative depth of pile foundation scour and time T are used as factors to obtain the pile foundation scour depth curve through a linear regression algorithm. Due to the influence of external sudden factors such as typhoons, the relative depth of pile foundation scour uploaded by the three-dimensional sonar is much greater than the pile foundation scour depth curve obtained by the linear regression algorithm. At this time, a typhoon-type mutation curve is formed; each pile foundation has an attention counter. Every time a high-level height vector appears, the attention counter value will increase. The pile foundation attention is classified and aggregated through the K-means algorithm to obtain the number of pile foundations with the highest attention. If the number of pile foundations with high attention accounts for the total number of pile foundations of the bridge, it means that the bridge safety attention is relatively high.

[0145] Specifically, the function implementation steps of the AI ​​artificial intelligence server also include: a prediction method for the impact of typhoons and earthquakes on pile foundation depth based on a fully connected neural network, see Fig.10 As shown, the specific steps are as follows:

[0146] (1) Obtain the data set required by the network: first obtain the typhoon intensity, wind direction, and earthquake intensity data within the time period, and then combine the pile foundation number, pile foundation type, and current depth of each pile foundation, plus the depth change obtained by subtracting the previous reported depth from the current depth, to obtain a seven-tuple {pile foundation number, pile foundation type, typhoon intensity, wind direction, earthquake intensity, current depth, depth change}. Take the first six items as the network input data X, and the last item is the network output value label y0. Sample the above sample data of all pile foundations and repeat this operation to obtain a large number of data sets, which are divided into training sets and test sets for network training and testing;

[0147] (2) Constructing a fully connected neural network: The fully connected neural network designed in this paper is a three-layer network, and each layer is fully connected. The specific network construction framework is as follows: X = (x1, ..., x n ) is the actual input vector of the first input layer, n = 6 is the number of neurons in the input layer; H = (h1, ..., h m ) is the output vector of the second hidden layer, m = 64 is the number of neurons in the hidden layer; y is the actual output result of the third output layer, representing the predicted depth change value. The parameters of the fully connected neural network are as follows: A = (a ij ) n×m is the weight matrix from the input layer to the hidden layer, where a ij The weight parameter value from the ith neuron to the jth neuron in the hidden layer; B = (b1, ..., b m ) is the weight vector from the hidden layer to the output layer, where b j is the weight parameter value from the jth neuron in the hidden layer to the output neuron; c = (c1, ..., c m ) is the bias vector of the hidden layer; d is the bias parameter value of the output layer; ReLU function is the activation function of the hidden layer; Sigmoid function is the activation function of the output layer; the maximum number of iterative training is max_epochs; the error threshold for stopping training is ε; the label of the network is the actual depth change value y0;

[0148] (3) Data normalization: normalize the data set obtained in step 1 and the actual depth change value y0, and limit the actual input data value of the input layer and the label data of the output layer to between 0 and 1;

[0149] (4) Network initialization: Initialize the following parameters: weight matrix A from input layer to hidden layer; weight vector B from hidden layer to output layer; bias vector C of hidden layer; bias parameter value d of output layer; set learning rate to 1e -4 ;

[0150] (5) Calculate the output of the hidden layer. The calculation method is as follows In the formula, R(·) is the activation function of the hidden layer. The ReLU function is selected as the activation function of the hidden layer, and its function definition is R(t)=max(0, t): In the formula, t represents the input of the activation function, and R(t) represents the output of the activation function;

[0151] (6) Calculate the actual output of the output layer. The calculation method is as follows In the formula, P(·) is the activation function of the output layer. Sigmoid function is selected as the activation function of the output layer, and its function definition is: In the formula, z represents the input of the activation function, P(z) represents the output of the activation function, and the Sigmoid function limits the output to the range of 0 to 1;

[0152] (7) The mean square error between the predicted depth change y and the actual depth change y0 is used as the loss function for training the fully connected neural network, which is expressed as L(y, y0) = (y-y0) 2 ;

[0153] (8) Update the parameters of the fully connected neural network according to the adaptive moment estimation algorithm, and set the learning rate to 1e -4 , the maximum number of iterative training times max_epochs is 1000, and back propagation is performed through the adaptive estimation algorithm to update the weight matrix A from the input layer to the hidden layer, the weight vector B from the hidden layer to the output layer, the bias vector C of the hidden layer, and the bias parameter value d of the output layer;

[0154] (9) Determine whether the training process is over. The judgment is as follows: if |L(y, y0)|>ε, return to step 4 and continue training and solving; if |L(y, y0)|≤ε, end the training process; if the current number of training times reaches the maximum number of training times max_epochs, end the training process;

[0155] (10) Determine the updated network parameters and save the network parameters after the training, that is, determine the weight matrix A from the input layer to the hidden layer, the weight vector B from the hidden layer to the output layer, the bias vector C of the hidden layer, and the bias parameter value d of the output layer; obtain the trained fully connected neural network according to the above steps, and obtain the depth change value y of the test data set through the updated network.

[0156] The method for predicting the influence of typhoon and earthquake on pile foundation depth based on fully connected neural network adopted in this embodiment uses fully connected neural network to complete the prediction of pile foundation depth change according to typhoon intensity, wind direction and earthquake intensity data in a data-driven machine learning manner, and reduces the problem to a regression problem. The main processing process is: in the model training stage, features are extracted from the pile foundation number, location, current depth and depth change sent by the pile foundation, and the historical data of typhoon intensity, wind direction and earthquake intensity sent by the meteorological station, a fully connected neural network is trained, and the gradient descent algorithm is executed to find the optimal parameters; in the model reasoning stage, the fully connected neural network that has been trained is used to predict the change of pile foundation depth. It can be seen from the test that the present invention has higher accuracy than the traditional method for predicting the influence of typhoon on pile foundation depth.

[0157] Embodiment 3

[0158] Based on the same concept, the present invention provides a three-dimensional sonar offshore pile foundation online monitoring device, comprising:

[0159] A transmitting module is used to transmit highly directional sound wave signals to the seabed / river bottom based on 3D sonar;

[0160] A judgment module is used to judge whether the acoustic wave signal encounters an obstacle. When the obstacle encountered is the seabed / river bottom, the echo signal is reflected and a detection threshold of the seabed signal is obtained. Based on the detection threshold, the relative distance between the seabed / river bottom and the 3D sonar is obtained, and the distance data from the acoustic wave emission point to the acoustic wave arrival end point is obtained;

[0161] A calculation module is used to obtain the arrival time and arrival direction of the current echo signal at the same time to obtain the beam pitch angle and azimuth corresponding to the current echo signal when the distance data is obtained; estimate the target position parameters of the echo based on the obtained arrival time and arrival direction of the current echo signal and based on a joint estimation algorithm of TOA and DOA; and calculate the target depth data through trigonometric functions in combination with the target position parameters.

[0162] The implementation principles of the above-mentioned transmitting module, judging module and calculating module have been described in the above-mentioned embodiments and will not be repeated here.

[0163] It should be noted that the division of the modules in the embodiment of the device / system is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or physically separated. Moreover, these modules can be implemented in the form of software calling through processing elements; or in the form of hardware; or some units can be implemented in the form of software calling through processing elements, and some units can be implemented in the form of hardware.

[0164] Embodiment 3

[0165] Based on the same concept, in some embodiments of the present application, an electronic device is also provided. The electronic device includes a memory and a processor, wherein the memory is used to store a processing program, and the processor executes the processing program according to instructions. When the processor executes the processing program, the three-dimensional sonar offshore pile foundation online monitoring method in the above-mentioned embodiment is implemented.

[0166] In some embodiments of the present application, a readable storage medium is also provided, which can be a non-volatile readable storage medium or a volatile readable storage medium. The readable storage medium stores instructions, and when the instructions are executed on a computer, an electronic device including the readable storage medium executes the aforementioned three-dimensional sonar offshore pile foundation online monitoring method.

[0167] It is understandable that, for the aforementioned three-dimensional sonar offshore pile foundation online monitoring method, if it is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially 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, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0168] Computer readable storage media may include data signals propagated in baseband or as part of a carrier wave, wherein readable program codes are carried. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program codes contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0169] The program code involved in executing the technical solution disclosed in the present application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and also conventional procedural programming languages ​​such as C language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A three-dimensional sonar offshore pile foundation online monitoring method, characterized in that: Used in 3D sonar offshore pile foundation online monitoring system, including: Based on three-dimensional sonar, highly directional sound wave signals are emitted to the seabed; Determine whether the acoustic wave signal encounters an obstacle, and when the obstacle encountered is the seabed, reflect the echo signal and obtain a detection threshold of the seabed signal, obtain the relative distance between the seabed and the three-dimensional sonar based on the detection threshold, and obtain the distance data from the acoustic wave emission point to the acoustic wave arrival end point; When the distance data is obtained, the arrival time and arrival direction of the current echo signal are simultaneously obtained to obtain the beam pitch angle and azimuth angle corresponding to the current echo signal; According to the acquired arrival time and arrival direction of the current echo signal, the target position parameters of the echo are estimated based on the joint estimation algorithm of TOA and DOA; Calculating target depth data by trigonometric function in combination with the target position parameters; The target position parameters of the echo estimated by the joint estimation algorithm based on TOA and DOA include: The joint weighted time average and azimuth deviation indication are used to obtain the precise arrival time of the echo signal and calculate the depth data at the current incident angle. The specific steps are as follows: When the incident angle is less than the preset angle threshold, the weighted time averaging algorithm is used to obtain the amplitude sequence corresponding to the echo signal, and the sampling points exceeding the threshold are selected from the received echo amplitude sequence, and the TOA value of the current echo signal is obtained by calculating the time-weighted average of the amplitude of the sampling points; When the incident angle of the echo is greater than or equal to the preset angle threshold, the BDI algorithm is used to obtain time slice samples. The received echo signals are arranged into a series of spatial grids according to the beam angle direction. For each time slice, the grids whose energy exceeds the threshold are selected to make an initial judgment on the number of echoes. The energy of a group of grids suspected of echoes is fit using a second-order polynomial to obtain the DOA value of the current echo signal.

2. The three-dimensional sonar offshore pile foundation online monitoring method according to claim 1 is characterized in that: The three-dimensional sonar comprises: A stepper motor controller is used to subdivide the drive according to the azimuth angle control requirements and the pitch angle control requirements, and adjust the rotation speed of the motor to achieve precise control of the azimuth angle and pitch angle; The transducer is connected to the stepper motor controller through a transmission shaft to form a 3D gimbal structure, which is used to control the stepper motor controller to drive the corresponding motor to rotate around the transmission shaft according to a preset azimuth angle range, drive the transducer to rotate for 360° scanning, and feedback the azimuth angle information through the sensor to complete the azimuth angle measurement. At the same time, according to the preset pitch angle, the stepper motor controller is controlled to drive the corresponding motor to rotate around the transmission shaft, drive the transducer to rotate for 90° scanning, and feedback the pitch angle information through the sensor to complete the pitch angle measurement; wherein, each subdivision angle of the stepper motor controller is accurate to 0.225°.

3. The three-dimensional sonar offshore pile foundation online monitoring method according to claim 1 is characterized in that: The internal circuit structure of the three-dimensional sonar includes a signal processor, a transmitting drive circuit, a data collector, a power amplifier, a signal conditioner, a transceiver conversion switch and a transducer. The signal processor controls the transmitting signal to be emitted from the transmitting drive circuit and the power amplifier, and the transmitting signal is transmitted to the transducer through the transceiver conversion switch. The transducer converts the electrical signal corresponding to the transmitting signal into an acoustic wave signal and transmits it to the seabed. The acoustic wave signal reflects an echo signal during propagation in the water. After being received by the transducer, the echo signal is transmitted to the signal processor through the transceiver conversion switch, the signal conditioner and the data collector for reading, analysis and processing, and finally generates the depth information of the target location. When the machine is turned on, under the control of the signal processor, the azimuth and elevation angles are reset to zero; Start sonar scanning according to the set working parameters, and divide the seabed sonar coverage area into multiple grid areas; The center of each grid area is mapped to the azimuth and elevation angle of the spherical coordinate system, and the corresponding motor is controlled by a stepper motor controller to align the sonar beam with the center of the grid area, and corresponding scanning and data acquisition operations are performed on the sonar beam.

4. The three-dimensional sonar offshore pile foundation online monitoring method according to claim 1 is characterized in that: The method of estimating the target location parameters of the echo based on the joint estimation algorithm of TOA and DOA further includes: When the angle of arrival of the echo signal is θ, the current beam angle and the front and rear beam angles are: β, β l and β r , and satisfy β-β l =β r -β=Δβ; The signal form of the three beams is expressed as: The echo amplitude of the previous beam angle is The echo amplitude at the current beam angle is The echo amplitude of the latter beam angle is Among them: a l , a, a r are the echo amplitudes of the previous beam, the current beam and the next beam respectively, and k is a constant; Subtract the logarithms of the echo amplitude of the previous beam angle, the echo amplitude of the current beam angle, and the echo amplitude of the next beam angle, and then sort them out to get the angle formula of the echo signal arrival as follows: Other:Δa r ln(a r )-ln(a),Δa l ln(a)-ln(a). l ), When the echo angle is known, the echo arrival angle DOA is calculated according to the angle formula of the echo signal arrival.

5. The three-dimensional sonar offshore pile foundation online monitoring method according to claim 1 is characterized in that: Calculating target depth data by trigonometric function in combination with the target position parameters includes: Establish a three-dimensional coordinate system, with the sonar probe as the coordinate origin O(0, 0, 0), confirm the direction of the transducer's azimuth angle α and pitch angle β' to transmit sound waves, and receive and detect echoes in this direction; When the sound wave encounters the target T (x0, y0, z0) and is scattered, the distance R between OT is calculated; The projection (x1, y1, z1) of the target T in the coordinate system is calculated according to the trigonometric function. The formula is as follows: x1=R*sin(β')*cos(α); y1=R*sin(β')*sin(α); z1=R*cos(β'); Among them, (x1, y1) is the projection position of the target T on the horizontal plane, and z1 is the depth of the target T.

6. The three-dimensional sonar offshore pile foundation online monitoring method according to claim 2 is characterized in that: The transducer includes a plurality of independent circular rings cut from the inside to the outside, and the areas of the plurality of distributed circular rings are obtained by cosine weighting, so that the energy applied to the plurality of circular rings is distributed according to the cosine weighting coefficient, thereby achieving the reduction of side lobes, wherein the cosine weighting formula is as follows: W(n)=1 / 2(1+cos(2πn / (N-1))), N is 16, n is 0, 1, 2, 3, 4, and the corresponding area ratios of S1 to S5 are as follows: S1:S2:S3:S4:S5=1:0.96:0.85:0.69:0.5, wherein W(n) refers to the cosine weighting function, and S1, S2, S3, S4, and S5 represent the areas corresponding to the plurality of independent circular rings cut from the inside to the outside of the transducer.

7. A three-dimensional sonar offshore pile foundation online monitoring system, characterized in that: To implement the three-dimensional sonar offshore pile foundation online monitoring method as described in any one of claims 1 to 6, the system comprises: The acquisition module includes a 3D sonar side-scan unit for scanning the local scour of each pile foundation, which is used to perform vector calculations on the acquired side-scan data and compress the monitoring data capacity through an efficient compression algorithm for 3D sonar data based on edge computing; The transmission module is used to combine the actual communication conditions on site and support wired network or wireless communication to achieve data transmission and analysis; A storage module is used to receive three-dimensional sonar data and extract three-dimensional sonar feature compression data for vector reconstruction, and automatically generate and store corresponding three-dimensional sonar monitoring dot cloud images; The service module includes a data cache server, a data docking server, an AI artificial intelligence server and a web application server. The data cache server is used to further optimize the vector structure data. Data docking server: docking with external system data to obtain more dimensional information and provide data sources of various dimensions for data mining; AI artificial intelligence server: The vector data generated by each 3D sonar side scan monitoring is quantified through the feature model to form regression vectors, low-level height vectors, and high-level height vectors. The corresponding vectors are selected as training samples for the target elements and input into the machine learning algorithm model for training to obtain quantitative, qualitative and trend analysis data related to pile foundation scour depth, scour type, morphology and correlation factors; Web application server: used to provide service support for visual data display; The application module is used to realize the aggregation, parsing, storage, analysis and sonar image display of local scour data of pile foundation through interface operation.

8. The three-dimensional sonar offshore pile foundation online monitoring system according to claim 7 is characterized in that: The steps for implementing the functions of the AI ​​artificial intelligence server include: Acquire pile foundation scour depth data, wherein the pile foundation scour depth data includes low-level height vector data and high-level height vector data; Based on the edge computing method, the low-level height vector and the high-level height vector are distinguished, and the pile foundation scouring depth data is compared with the preset depth threshold. If the current pile foundation scouring depth data is greater than the preset depth threshold, it is determined to be high-level height vector data; if the current pile foundation scouring depth data is less than or equal to the preset depth threshold, it is determined to be low-level height vector data, and the high-level height vector data is uploaded to the data queue in the order of priority, and the high-level height vector data is processed first and then the low-level height vector data; The first pile foundation scouring depth curve is obtained by using a linear regression algorithm with the pile foundation scouring relative depth and time T as factors. When affected by external factors, the second pile foundation scouring relative depth data uploaded by the three-dimensional sonar is greater than the first pile foundation scouring depth data obtained by the first pile foundation scouring depth curve. At this time, it is determined to be an exogenous mutation curve, and the external factors at least include typhoons. Each pile foundation is equipped with an attention counter. When a high-height vector appears, the corresponding attention counter value is increased to obtain the attention count data of each pile foundation. The attention count data of each pile foundation is classified and aggregated through the K-means algorithm to obtain the number of pile foundations with the highest attention. If the proportion of high-attention pile foundations to the total number of pile foundations of the bridge is greater than the preset attention threshold, it is judged that the bridge safety attention is relatively high.

9. An electronic device, characterized in that: include: A memory, the memory being used to store a processing program; A processor, wherein when executing the processing program, the processor implements the three-dimensional sonar offshore pile foundation online monitoring method as described in any one of claims 1 to 6.