Multi-source data fusion anti-wave performance evaluation and analysis system for seawall protection structure

Abstract

This invention relates to the field of water conservancy engineering safety monitoring technology, and discloses a multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures. The system includes a data acquisition subsystem and a data processing terminal. The acquisition subsystem acquires wave elevation, structural vibration, and pore water pressure data. The processing terminal performs data synchronization verification, dynamic benchmark construction, and multi-dimensional feature fusion to evaluate wave resistance performance. The method uses real-time tide levels and fuzzy membership functions to synthesize dynamic health benchmarks, eliminating nonlinear interference from water level changes. Through frequency and time domain dual-modal analysis, it calculates the transfer function drift reflecting stiffness changes and the hysteresis deviation reflecting permeability characteristics. This invention solves the problems of traditional methods being greatly affected by environmental conditions and difficulty in identifying internal hidden damage, achieving accurate differentiation between external revetment loosening and internal cavitation and piping, and improving the accuracy of health diagnosis of seawall protection structures.

Description

Technical Field

[0001] This invention relates to the field of water conservancy engineering safety monitoring technology, specifically to a multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures. Background Technology

[0002] As the first line of defense against storm surge disasters and to protect coastal life and property, seawall structures are constantly exposed to complex marine dynamic environments. Under the continuous cyclic loads of waves and tidal erosion, seawall structures are prone to damage such as loosening of the revetment blocks, interlocking failure, and loss of internal soil. If these damages are not detected and repaired in time, they can easily lead to overall structural instability or even seawall collapse under extreme sea conditions. Therefore, establishing an effective wave resistance performance assessment and analysis system to monitor the dynamic response characteristics of seawalls under wave impact in real time is of great engineering significance for early identification of structural defects and assessment of their safe service status.

[0003] Existing monitoring technologies for seawall protection structures mainly rely on vibration signal analysis. These technologies typically involve deploying accelerometers or vibration meters at key locations along the seawall to collect vibration response signals of the structure under wave impact. At the data processing level, existing technologies generally employ modal analysis or transfer function analysis. This involves calibrating the inherent frequency response function of the structure as a benchmark reference model during the initial construction phase or when the seawall is in good condition. In subsequent monitoring, the frequency response function of the real-time acquired signals is calculated and compared with the preset fixed benchmark model. The drift or change in correlation coefficient of the frequency response function is used to determine whether the structural stiffness has decreased, thereby inferring whether structural damage exists.

[0004] While existing technologies can detect structural anomalies to some extent through changes in vibration characteristics, they still have some shortcomings. First, existing monitoring methods typically set health benchmarks based on the assumption of a fixed water level, ignoring the nonlinear impact of tidal level changes on the structural dynamic characteristics. Seawall structures actually operate in a variable water level environment, and the rise and fall of water levels significantly alter the added mass of the water acting on the sloping seawall. Fluctuations in this added mass directly cause drifts in the structure's natural frequency and damping ratio. If only a single static benchmark is used for comparison, the normal physical parameter drifts caused by tidal changes are often misjudged as structural damage, leading to frequent false alarms from the system. Second, existing single vibration monitoring methods are unable to distinguish the specific physical mechanisms of damage. Loosening of the external revetment blocks and cavitation and piping formed by internal soil, although both lead to a decrease in the overall structural stiffness and a change in the frequency response function, have completely different failure mechanisms and emergency response methods. Relying solely on vibration frequency domain characteristics cannot capture the unique hydraulic permeability changes of internal cavities, making it difficult to effectively identify the more concealed and dangerous internal piping channels in the early stages, posing a risk of missing major safety hazards. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures. This system solves the problems that are easily caused by nonlinear changes in tide levels, leading to frequent false alarms, and that relying solely on single vibration monitoring makes it difficult to effectively identify and distinguish between hidden piping inside the seawall and damage to the external revetment.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of the present invention provides a multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures, which mainly consists of a data acquisition subsystem and a data processing terminal.

[0007] The data acquisition subsystem is used to acquire environmental load data and seawall physical response data, and includes an environmental monitoring unit, a structural response monitoring unit, and a hydraulic response monitoring unit. The environmental monitoring unit is equipped with wave radar and a tide gauge to acquire time-series signals of wave surface elevation and static water level height data, respectively. The structural response monitoring unit includes an array of vibration sensors deployed at key stress points of the seawall to acquire vibration acceleration signals and dynamic strain signals of the seawall structure under wave impact. The hydraulic response monitoring unit includes pore water pressure sensors buried at a specific depth inside the seawall to acquire fluid pressure response data of the media inside the seawall.

[0008] The data processing terminal is connected to the data acquisition subsystem via an industrial fieldbus or fiber optic network. Internally, it includes a signal synchronization module, a data verification module, a benchmark construction module, a feature analysis module, and a fusion diagnostic module. The signal synchronization module receives multi-source signals and calculates the physical lag time using a cross-correlation function, performing time-domain translation compensation on wave signals to achieve spatiotemporal alignment. The data verification module performs energy-gated data validity judgment and spatial consistency-based sensor fault elimination. The benchmark construction module maps the state subspace based on real-time tide data to synthesize a dynamic health benchmark under the current operating conditions. The feature analysis module performs dual-modal analysis in the frequency and time domains, calculating the transfer function drift and hysteresis deviation. The fusion diagnostic module maps the feature quantities to a two-dimensional fault feature plane for joint diagnosis and outputs the health status judgment result.

[0009] The second aspect of this invention provides a method for evaluating and analyzing the wave resistance performance of seawall protection structures using multi-source data fusion. This method is based on the aforementioned system and specifically includes multi-source data synchronous acquisition, data validity gating and quality verification, dynamic benchmark matching of tidal level modulation, joint analysis of frequency domain and time domain features, and multi-dimensional feature fusion diagnostic steps.

[0010] The data validity gating and quality verification steps include: calculating the root mean square energy value of the wave signal within the current time window. Calculate using the following formula: ; in, The length of the time window. This is the wave surface elevation signal after spatiotemporal synchronization. If... Below the preset minimum effective excitation threshold The current data is deemed invalid; only when The data is then deemed valid. Simultaneously, for redundant vibration sensors deployed on the same monitoring section, the validity of any two vibration sensors is calculated. and cross-relationships between : ; in, For signal covariance, and This represents the signal standard deviation. If the cross-correlation coefficient of a sensor is lower than a set threshold, it is discarded.

[0011] The joint analysis steps of frequency domain and time domain features include: utilizing the self-power spectral density of wave signals. and the cross-power spectral density of wave signals and structural vibration signals Calculate the current frequency response transfer function Subsequently, the Euclidean distance between the current frequency response transfer function and the synthesized reference frequency response function is calculated to obtain the transfer function drift. ; in, and To effectively analyze the upper and lower frequency limits of the frequency band, the timing of wave impacts on the seawall surface is extracted in the time domain. and the moment when pore water pressure changes abruptly Calculate the current hydraulic seepage lag time. The difference between the hysteresis time and the synthetic reference hysteresis time is calculated to obtain the hysteresis deviation. ; The multidimensional feature fusion diagnostic step includes: setting a stiffness drift threshold. and permeability deviation threshold The transfer function drift Hysteresis deviation The fault is mapped to a two-dimensional fault feature plane for partitioning and determination. The specific diagnostic logic is as follows: when and At that time, it was determined that the seawall structure had loose external revetment or interlocking failure; when and At that time, it was determined that there were cavities or piping channels inside the seawall structure; when and If this occurs, it is determined that there is a local seepage anomaly or a pore water pressure sensor malfunction. when and At that time, it was determined that the seawall structure was in a healthy state.

[0012] By introducing a dynamic benchmark construction mechanism based on tidal level modulation, the problem of false alarms caused by nonlinear changes in the dynamic characteristics of seawall structures with water level was solved. By integrating the dual-mode characteristics of structural vibration response and pore water pressure response, a two-dimensional fault characteristic plane was constructed, which enabled effective differentiation between external revetment loosening and internal void piping, thus improving the accuracy and robustness of the wave resistance performance assessment of seawall protection structures.

[0013] This invention provides a multi-source data fusion system for evaluating and analyzing the wave resistance performance of seawall protection structures. It offers the following advantages: 1. This invention effectively solves the problem of false alarms caused by changes in environmental conditions by constructing a dynamic health benchmark based on tidal level modulation. Existing technologies typically use fixed benchmarks, ignoring the nonlinear influence of the added mass of water caused by tidal level changes on the frequency characteristics of the structure. This invention maps the real-time tidal level to the state subspace and uses fuzzy membership functions to weight and synthesize pre-calibrated benchmark parameters to generate a dynamic benchmark that matches the current water level in real time. This mechanism can automatically cancel the interference of tidal fluctuations on the monitoring signal and ensure that the drift of characteristic parameters only reflects the physical damage of the structure itself, thereby significantly reducing false alarms triggered by changes in environmental boundary conditions.

[0014] 2. This invention utilizes the fusion diagnosis of frequency and time domain dual-modal features to achieve accurate classification of external damage and internal hidden dangers of seawalls. Traditional single vibration monitoring is difficult to distinguish the specific causes of structural stiffness reduction. This invention, while obtaining the transfer function drift quantity reflecting stiffness changes, introduces the hysteresis deviation quantity reflecting hydraulic seepage characteristics and maps both to a two-dimensional fault feature plane. By analyzing the distribution logic of these two feature quantities, the system can accurately distinguish between external revetment loosening and internal voids or piping, making up for the deficiency of conventional methods in identifying hidden diseases inside the dike.

[0015] 3. This invention establishes a data quality control system that includes energy gating and spatial consistency verification, improving the reliability of the evaluation results. For the field environment characterized by large sea state fluctuations and easily damaged sensors, the system sets a minimum effective excitation threshold, performing feature extraction only when wave energy is sufficient to excite an effective modal response of the structure. This avoids calculation errors caused by minor environmental noise. Simultaneously, by utilizing the redundancy characteristics of array sensors for spatial cross-correlation analysis, it can identify and eliminate faulty sensors with abnormal data in real time, preventing inferior data from entering subsequent diagnostic processes and ensuring the operational stability and data reliability of the long-term online monitoring system. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the system architecture of the present invention; Figure 2 This is a flowchart of the overall evaluation and analysis method of the present invention; Figure 3 This is a flowchart of the data validity verification and dynamic benchmark construction process of the present invention; Figure 4 This is a flowchart of the feature analysis and fusion diagnosis process of the present invention; Figure 5 This is a comparison chart of the frequency response function drift of the present invention; Figure 6 This is a comparison chart of the frequency response function drift under different tidal conditions according to the present invention.

[0017] The system comprises: 100, Data Acquisition Subsystem; 110, Environmental Monitoring Unit; 111, Wave Radar; 112, Tide Gauge; 120, Structural Response Monitoring Unit; 121, Vibration Sensor; 130, Hydraulic Response Monitoring Unit; 131, Pore Water Pressure Sensor; 200, Data Processing Terminal; 210, Signal Synchronization Module; 220, Data Verification Module; 230, Benchmark Construction Module; 240, Feature Analysis Module; and 250, Fusion Diagnosis Module. Detailed Implementation The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] See attached document Figure 1 , Figure 1This is a schematic diagram of a multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures according to an embodiment of the present invention. The system mainly consists of a data acquisition subsystem 100 and a data processing terminal 200. The data acquisition subsystem 100 is responsible for acquiring environmental load data and seawall physical response data, while the data processing terminal 200 is responsible for spatiotemporal alignment, feature extraction, and fusion diagnosis of the data.

[0019] The data acquisition subsystem 100 includes an environmental monitoring unit 110, a structural response monitoring unit 120, and a hydraulic response monitoring unit 130.

[0020] The environmental monitoring unit 110 is used to collect wave energy data and environmental boundary condition data. The environmental monitoring unit 110 further includes a wave radar 111 and a tide gauge 112. The wave radar 111 is installed on a fixed tower or sea surface observation platform in front of the seawall's wave-facing side, and its transmitted beam covers a specific water area in front of the seawall, collecting time-series signals of wave surface elevation. The tide gauge 112 is installed in a still water well or on the backwater side of the seawall to obtain still water level data.

[0021] The structural response monitoring unit 120 is used to collect solid vibration response data of the seawall structure under wave impact. The structural response monitoring unit 120 includes an array of vibration sensors 121 deployed at key stress-bearing parts of the seawall. The key stress-bearing parts include the surface of the revetment block, the top of the wave wall, and the connection of the caisson. The vibration sensors 121 collect vibration acceleration signals and dynamic strain signals. On the same monitoring section, adjacent vibration sensors 121 are configured with redundant nodes to maintain a preset spatial distance.

[0022] The hydraulic response monitoring unit 130 is used to collect fluid pressure response data of the medium inside the seawall under the action of waves. The hydraulic response monitoring unit 130 includes a pore water pressure sensor 131 buried at a specific depth inside the seawall. The pore water pressure sensor 131 is buried in the backfill area behind the caisson, the prism riprap area, or the inside of the geotextile. The pore water pressure sensor 131 has the dynamic response capability to capture millisecond-level transient pressure fluctuations and collect pore water pressure change signals.

[0023] The data processing terminal 200 is connected to the data acquisition subsystem 100 via an industrial fieldbus or fiber optic network. The data processing terminal 200 includes a signal synchronization module 210, a data verification module 220, a benchmark construction module 230, a feature analysis module 240, and a fusion diagnostic module 250.

[0024] See attached document Figure 2 , Figure 2 This is a flowchart of an evaluation and analysis method according to an embodiment of the present invention. The present invention provides a multi-source data fusion wave resistance performance evaluation and analysis method for seawall protection structures, comprising the following steps: S1, the signal synchronization module 210 receives wave signals from wave radar 111, vibration signals from vibration sensor 121 and pore pressure signals from pore water pressure sensor 131. The signal synchronization module 210 uses the cross-correlation function to calculate the physical lag time of wave propagation from the monitoring point to the seawall surface, performs time-domain translation compensation on the wave signals, and realizes the spatiotemporal alignment of the input and output signals.

[0025] S2, the data verification module 220 calculates the root mean square energy value of the wave signal within the current time window and compares it with a preset minimum effective excitation threshold. If the root mean square energy value is lower than the minimum effective excitation threshold, the evaluation state of the previous moment is maintained; if the root mean square energy value is higher than or equal to the minimum effective excitation threshold, subsequent processing begins. The data verification module 220 also calculates the cross-correlation coefficient between adjacent vibration sensors 121 on the same cross-section and discards sensor data with correlation coefficients lower than the fault determination threshold.

[0026] S3, the benchmark construction module 230 receives real-time tide data and maps it to a preset tide state subspace. The benchmark construction module 230 uses a fuzzy membership function to calculate the weighted average of the current tide level to the benchmark models in adjacent subspaces, synthesizing a dynamic health benchmark under the current operating conditions. The dynamic health benchmark includes the synthesized benchmark frequency response function and the synthesized benchmark lag time.

[0027] S4, Feature Analysis Module 240 performs dual-modal analysis in the frequency and time domains. It calculates the actual frequency response function between the wave signal and the vibration signal, and the Euclidean distance between the actual frequency response function and the synthesized reference frequency response function, obtaining the transfer function drift. Feature Analysis Module 240 extracts the time difference between the wave impact moment and the moment of sudden change in pore water pressure, calculates the actual hydraulic seepage hysteresis time, and the difference between the actual hydraulic seepage hysteresis time and the synthesized reference hysteresis time, obtaining the hysteresis deviation.

[0028] S5, the fusion diagnostic module 250 performs joint diagnosis based on the transfer function drift and hysteresis deviation. The fusion diagnostic module 250 maps these two characteristic quantities to a two-dimensional fault characteristic plane for classification and judgment: if only the transfer function drift is abnormal, it is judged as loosening of the external retaining; if the transfer function drift is abnormal and the hysteresis deviation increases negatively, it is judged as the formation of internal voids or piping channels; if only the hysteresis deviation is abnormal, it is judged as local seepage abnormality or sensor failure. The fusion diagnostic module 250 outputs the evaluation results and maintenance suggestions.

[0029] See attached document Figure 3 , Figure 3This is a schematic diagram of a data validity verification and dynamic benchmark construction process according to an embodiment of the present invention. The present invention provides a method for nonlinear interference removal and dynamic benchmark construction, mainly executed by a data verification module 220 and a benchmark construction module 230, including the following steps: S21, the data verification module 220 acquires the wave signal after spatiotemporal alignment and calculates the root mean square energy value of the wave signal within the current time window. The calculation formula is as follows: ; in, The length of the time window. This is the wave surface elevation signal after spatiotemporal synchronization.

[0030] S22, the data verification module 220 will verify the root mean square energy value. With the preset minimum effective excitation threshold When comparing, the root mean square energy value Less than the minimum effective excitation threshold When the wave energy is insufficient to excite an effective mode, a hold command is output. The system pauses the transfer function update for the current time window and maintains the evaluation result from the previous effective time step. When the root mean square energy value... Greater than or equal to the minimum effective excitation threshold When the current data is valid, it is determined that the data is valid; a multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures.

[0031] S23, the data verification module 220 performs a spatial consistency verification of the sensor array. For redundant vibration sensors 121 deployed on the same monitoring section, it calculates the consistency of any two vibration sensors. and cross-relationships between The calculation formula is as follows: ; in, For signal covariance, and The standard deviation of the signal is defined as follows: if the cross-correlation coefficient between a vibration sensor 121 and other vibration sensors 121 in the group is lower than the fault determination threshold... Furthermore, the cross-correlation coefficients among other vibration sensors 121 within the group are higher than the fault determination threshold. The vibration sensor 121 was determined to be faulty and its signal was removed. S31, the baseline construction module 230 receives real-time tide data. Calculate the average tide level within the current time window. Based on pre-stored Tide level state subspace The current average tide level is calculated using fuzzy membership functions. For each state subspace Membership weights The fuzzy membership function satisfies the normalization condition: ; S32, the benchmark construction module 230 performs weighted synthesis of benchmark parameters in adjacent state subspaces according to membership weights to construct a dynamic health benchmark under the current operating condition and synthesize the benchmark frequency response function. The calculation formula is as follows: ; Synthetic reference lag time The calculation formula is as follows: ; in, and The first The pre-calibrated benchmark parameters corresponding to each state subspace are transmitted from the benchmark construction module 230 to the subsequent feature analysis module 240.

[0032] A multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures (see attached document) Figure 4 , Figure 4 This is a schematic diagram of a feature analysis and fusion diagnosis process according to an embodiment of the present invention. The present invention provides a multi-dimensional feature extraction and fluid-structure interaction fusion diagnosis method, mainly executed by a feature analysis module 240 and a fusion diagnosis module 250, including the following steps: S41, the feature analysis module 240 receives the spatiotemporally aligned wave signal and structural vibration signal, performs a fast Fourier transform on the wave signal and structural vibration signal, and calculates the self-power spectral density of the wave signal. and the cross-power spectral density of wave signals and structural vibration signals The feature analysis module 240 calculates the current frequency response transfer function using the self-power spectral density and cross-power spectral density. The calculation formula is as follows: ; Feature analysis module 240 receives the synthesized reference frequency response function from reference construction module 230. Calculate the current frequency response transfer function With the synthesized reference frequency response function The Euclidean distance between them yields the transfer function drift. The calculation formula is as follows: ; in, and To effectively analyze the upper and lower limits of the frequency band; S42, Feature analysis module 240 extracts the time of wave impact on the seawall surface in the time domain. and the moment when pore water pressure changes abruptly Calculate the current hydraulic seepage lag time. The calculation formula is: Feature analysis module 240 receives the synthesized benchmark lag time from benchmark construction module 230. Calculate the current hydraulic seepage lag time. Synthetic reference lag time The difference is used to obtain the hysteresis bias. The calculation formula is as follows: ; S51, the fusion diagnostic module 250 receives the transfer function drift. Hysteresis deviation Set stiffness drift threshold and permeability deviation threshold The transfer function drift Hysteresis deviation Mapping to a two-dimensional fault feature plane for partitioning and determination; S52, the integrated diagnostic module 250 executes diagnostic logic: when the transfer function drift... Greater than the stiffness drift threshold And hysteresis deviation The absolute value of the multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures is less than or equal to the permeability deviation threshold. When the seawall structure is found to have loose external cladding or interlocking failure, the transfer function drift is determined to be... Greater than the stiffness drift threshold And hysteresis deviation Less than the negative permeability deviation threshold When the seawall structure is found to have cavities or piping channels, it is determined that there are such cavities or piping channels within it; when the transfer function drift is... Less than or equal to the stiffness drift threshold And hysteresis deviation The absolute value is greater than the permeability deviation threshold. When this occurs, it is determined that there is a local seepage anomaly or a malfunction in pore water pressure sensor 131; when the transfer function drift... A multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures is required to have a stiffness drift threshold of less than or equal to the threshold value. And hysteresis deviation The absolute value is less than or equal to the permeability deviation threshold. At that time, it was determined that the multi-source data fusion wave resistance performance evaluation and analysis system for the seawall structure facing the seawall protection structure was in a healthy state; S53, the fusion diagnostic module 250 generates a wave resistance performance evaluation report and maintenance instructions based on the judgment results, and outputs them to the external monitoring platform through the data processing terminal 200 to the multi-source data fusion wave resistance performance evaluation and analysis system of the seawall protection structure.

[0033] To verify the effectiveness of the system of the present invention, it was deployed and tested on-site in a standard coastal seawall demonstration section (from chainage K2+100 to K2+300).

[0034] System deployment parameters: Environmental monitoring unit 110: Employs a 24GHz K-band wave radar, installed on an observation tower 50 meters from the seawall's edge, with a sampling frequency set to 4Hz. The tide gauge is a radar-type water level gauge, outputting average tide data every 10 minutes.

[0035] Structural response monitoring unit 120: A three-axis MEMS accelerometer node is installed every 10 meters along the axial direction on the top of the caisson wave wall, with a sampling frequency set to 100Hz and a sensitivity of 2000mV / g.

[0036] Hydraulic response monitoring unit 130: A high-frequency pore water pressure sensor is pre-embedded in the backfill sand layer behind the caisson, at a depth of 3 meters below the top surface of the seawall. The response time is less than 5ms and the range is 0-200kPa.

[0037] Typical operating condition verification process: A typical storm surge event (lasting 48 hours) was selected as the test window. During this period, the seawall experienced dramatic changes from low tide to high tide (tidal range of 4.5 meters), accompanied by significant wave impact (significant wave height). The largest reaches 2.8 meters.

[0038] Benchmark construction phase: During the calm period before the storm surge, the system automatically collects structural response data at different tide levels, divides the tide level interval [0m, 5m] into 10 state subspaces (step size 0.5m), and establishes a corresponding tide level-frequency response benchmark library.

[0039] Real-time monitoring phase: T=12h: The real-time tide level is 4.2 meters. The system identifies this tide level as being in the 9th subspace and automatically calls the reference frequency response function under high water levels. Although the structure's natural frequency decreases by about 5% compared to the dry state due to the added mass effect of the water, the calculated transfer function drift is reduced due to the synchronous adjustment of the reference model. Still below the stiffness drift threshold The system correctly determined it to be in a "healthy state," thus avoiding false alarms.

[0040] T=36h: In order to verify the diagnostic logic, artificial high-pressure flushing was carried out at the test section K2+150 through a preset water injection pipeline to simulate the loss of internal soil (initial stage of piping).

[0041] At this time, the vibration signal on the structural surface did not change significantly, and the transfer function drift was... It remains at a low level. However, the internal pore water pressure sensor 131 detects a significant increase in the pressure wave propagation speed, resulting in a calculated hysteresis deviation. The negative increase (from the normal 1.2 seconds to 0.3 seconds) exceeded the negative osmotic deviation threshold. .

[0042] Based on the logic in step S52, the fusion diagnostic module 250 accurately captures " Normal and The combination of negative anomalies output an alarm indicating "formation of internal voids or piping channels," verifying the system's ability to identify hidden damage.

[0043] Experimental verification and effect comparison: To further quantify the advantages of the present invention over existing technologies, a fluid-structure interaction numerical simulation environment was constructed using software, and a comparative analysis was conducted in conjunction with field measurement data.

[0044] 1. Comparison object settings: Existing technical solution A (traditional threshold method): only monitors the peak value of vibration acceleration, and alarms when the peak value exceeds a fixed threshold. It does not consider the influence of tide level and does not integrate hydraulic data.

[0045] Existing technical solution B (static benchmark method): adopts frequency response function analysis, but uses a single mean tide level benchmark model and does not perform dynamic subspace matching.

[0046] The present invention employs a dynamic reference based on tidal level modulation and a fusion diagnosis of fluid-structure interaction features.

[0047] 2. Analysis of simulation experiment results: See attached document Figure 5 , Figure 5 This is a comparison chart of the frequency response function drift of the embodiments of the present invention and the prior art under different tidal conditions.

[0048] The horizontal axis in the figure represents the time axis (covering the process from low tide to high tide), and the vertical axis represents the frequency response function drift (normalized value).

[0049] Curve analysis: Under normal, undamaged operating conditions, the dynamic characteristics of the seawall structure undergo natural nonlinear drift as the tide level rises. The drift of prior art B (shown by the dashed line) increases significantly with rising tide level. The alarm threshold (red horizontal line) was mistakenly exceeded at the (high tide) level, generating a false alarm.

[0050] The present invention (shown by solid lines) introduces tidal weight correction in real time through the "benchmark construction module 230". The actual calculated drift amount always fluctuates near the zero line and is not affected by water level changes, which proves the superior ability of dynamic benchmark technology to eliminate various nonlinear environmental noises.

[0051] See attached document Figure 6 , Figure 6 This is a diagram illustrating the diagnostic clustering effect of an embodiment of the present invention on a two-dimensional fault feature plane. The horizontal axis in the diagram represents the transfer function drift. The vertical axis represents the hysteresis deviation. The figure shows the distribution of three typical sample points: Region I (Health Domain): Sample points are concentrated near the origin, corresponding to a healthy state.

[0052] Region II (External Damage Domain): Sample points are distributed along the positive horizontal axis. (Increase), but the change in the vertical axis is not obvious, corresponding to the loosening of the protective surface.

[0053] Region III (Internal Damage Domain): Sample points are significantly shifted in the negative direction along the vertical axis. ),and The changes were minor.

[0054] Effect Description: Existing technologies A and B can only perform projection judgment on a one-dimensional coordinate axis (horizontal axis), therefore they cannot distinguish between "Region I" and "Region III," leading to false negatives in the detection of internal piping-type defects. This invention introduces... The dimension successfully separated hidden internal damage from healthy samples.

[0055] 3. Performance Indicator Comparison Table: The statistical results based on 500 test samples (including 300 healthy samples, 100 samples with external injuries, and 100 samples with internal injuries) are shown in the table below: Experimental data show that this invention effectively solves the false alarm problem under high water conditions through "dynamic reference matching based on tidal level modulation," reducing the false alarm rate by 16.7 percentage points. Simultaneously, through "joint analysis of energy transfer and phase locking," it utilizes hydraulic hysteresis characteristics to compensate for the insensitivity of single vibration monitoring to internal damage, significantly reducing the missed alarm rate for internal cavities from 76.0% to 4.2%. This solution demonstrates significantly better robustness and accuracy than existing technologies in complex marine environments, possessing extremely high engineering practical value.

[0056] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures, characterized in that, Includes a data acquisition subsystem and a data processing terminal; The data acquisition subsystem is used to acquire wave energy data, solid vibration response data of the seawall structure, and fluid pressure response data of the medium inside the seawall. The data processing terminal includes a signal synchronization module, a data verification module, a benchmark construction module, a feature analysis module, and a fusion diagnosis module. The benchmark construction module is used to receive real-time tide data and map the real-time tide data to a preset tide state subspace, and use fuzzy membership functions to synthesize a dynamic health benchmark that includes a synthesized benchmark frequency response function and a synthesized benchmark lag time. The feature analysis module is used to perform dual-modal analysis in the frequency and time domains to calculate the transfer function drift that reflects changes in structural stiffness and the hysteresis deviation that reflects hydraulic permeability characteristics. The fusion diagnostic module is used to map the transfer function drift and the hysteresis deviation to a two-dimensional fault feature plane for joint diagnosis, and output the health status judgment result of the seawall structure.

2. The multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures according to claim 1, characterized in that, The data acquisition subsystem includes: an environmental monitoring unit, comprising wave radar and tide gauge, for acquiring time-series signals of wave surface elevation and static water level data; a structural response monitoring unit, comprising arrayed vibration sensors deployed at key stress points of the seawall, for acquiring vibration acceleration signals and dynamic strain signals; and a hydraulic response monitoring unit, comprising pore water pressure sensors buried at a specific depth inside the seawall dam, for acquiring pore water pressure change signals.

3. The multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures according to claim 1, characterized in that, The signal synchronization module is used to receive wave signals from the wave radar, vibration signals from the vibration sensor, and pore pressure signals from the pore water pressure sensor. It uses a cross-correlation function to calculate the physical lag time of wave propagation from the monitoring point to the seawall surface, and performs time-domain translation compensation on the wave signals to achieve spatiotemporal alignment between the input and output signals.

4. The multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures according to claim 1, characterized in that, The data verification module is used to calculate the root mean square energy value of the wave signal within the current time window and compare the root mean square energy value with a preset minimum effective excitation threshold. When the root mean square energy value is lower than the minimum effective excitation threshold, the system maintains the evaluation state of the previous moment. When the root mean square energy value is higher than or equal to the minimum effective excitation threshold, the current data is determined to be valid and the system enters the subsequent processing flow.

5. The multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures according to claim 1, characterized in that, The data verification module is also used to perform spatial consistency verification of the sensor array and calculate the cross-correlation coefficient between any two vibration sensors on the same monitoring section. If the cross-correlation coefficient between a vibration sensor and other vibration sensors in the group is lower than the fault judgment threshold, and the cross-correlation coefficient between other vibration sensors in the group is higher than the fault judgment threshold, then the vibration sensor is judged to be faulty and its signal is removed.

6. The multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures according to claim 1, characterized in that, The specific method by which the benchmark construction module synthesizes the dynamic health benchmark is as follows: calculate the membership weight of the average tide level within the current time window to each pre-stored tide level state subspace, and perform weighted synthesis of the pre-calibrated benchmark parameters corresponding to adjacent state subspaces based on the membership weight; the synthesized benchmark frequency response function is obtained by weighted summation of the benchmark frequency response functions of each state subspace, and the synthesized benchmark lag time is obtained by weighted summation of the benchmark lag times of each state subspace.

7. The multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures according to claim 1, characterized in that, The feature analysis module calculates the transfer function drift by first using the self-power spectral density of the wave signal and the cross-power spectral density of the wave signal and the structural vibration signal to calculate the current frequency response transfer function, and then calculating the Euclidean distance between the current frequency response transfer function and the synthetic reference frequency response function generated by the reference construction module within the effective analysis frequency band to obtain the transfer function drift.

8. The multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures according to claim 1, characterized in that, The feature analysis module calculates the hysteresis deviation by extracting the moment when the wave impacts the seawall surface and the moment when the pore water pressure changes abruptly in the time domain, and calculating the difference between the two to obtain the current hydraulic infiltration hysteresis time. The difference between the current hydraulic seepage hysteresis time and the synthetic benchmark hysteresis time generated by the benchmark construction module is calculated to obtain the hysteresis deviation.

9. The multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures according to claim 1, characterized in that, The fusion diagnostic module sets stiffness drift threshold and permeability deviation threshold, and executes the following diagnostic logic: When the transfer function drift is greater than the stiffness drift threshold and the absolute value of the hysteresis deviation is less than or equal to the permeability deviation threshold, it is determined that the seawall structure has external cladding loosening or interlocking failure. When the transfer function drift is greater than the stiffness drift threshold and the hysteresis deviation is less than the negative permeability deviation threshold, it is determined that there are cavities or piping channels inside the seawall structure. When the transfer function drift is less than or equal to the stiffness drift threshold, and the absolute value of the hysteresis deviation is greater than the permeability deviation threshold, it is determined that there is a local seepage anomaly or a pore water pressure sensor malfunction.

10. The multi-source data fusion wave resistance performance evaluation and analysis system for seawall protection structures according to claim 1, characterized in that, The data processing terminal is connected to the data acquisition subsystem via an industrial fieldbus or fiber optic network. The fusion diagnostic module is used to generate a wave resistance performance assessment report and maintenance instructions based on the judgment results, and outputs them to an external monitoring platform through the data processing terminal.

Images