A method and system for stress distribution analysis of permeable breakwaters under different wave conditions

Abstract

This invention proposes a method and system for stress distribution analysis of permeable breakwaters under different wave conditions, relating to the field of coastal engineering technology. The method includes: real-time synchronous acquisition of wave information and structural response information using sensors deployed at structural components and water-adjacent areas, generating a synchronous wave-structure response dataset; based on this dataset, analyzing the correspondence between wave characteristics and structural strain data, grouping by wave phase time sequence and constructing a structural stress topology map to generate a wave-stress correlation map; determining the correlation map region based on the characteristic information of the current wave condition and the map; and analyzing the structural stress distribution information contained in this region to obtain the stress distribution of the permeable breakwater under the current wave condition. This achieves a fine mapping between wave dynamic characteristics and structural stress distribution, improving analysis accuracy and efficiency.

Description

Technical Field

[0001] This invention relates to the field of coastal engineering technology, and in particular to a method and system for analyzing the stress distribution of permeable breakwaters under different wave conditions. Background Technology

[0002] As a common coastal protection structure, perforated breakwaters, while dissipating waves, also lead to more complex wave-structure interactions. Under different wave conditions (such as different combinations of wave height, period, direction, and phase), the stress distribution inside the structure exhibits dynamic and non-uniform characteristics. Traditional stress analysis methods based on static or typical wave conditions are insufficient to accurately reflect the true stress state. Current technologies typically employ finite element simulation combined with empirical formulas for stress estimation. However, this method relies on assumed load distributions and fails to fully utilize actual monitoring data. In particular, it lacks a detailed characterization of the correlation between wave dynamic phase and structural response time sequence, resulting in discrepancies between the analysis results and actual conditions. Consequently, it cannot provide accurate and real-time basis for structural safety assessment and optimization.

[0003] Therefore, there is an urgent need for an analytical method that can integrate real-time wave information and structural response data, and establish a direct and precise mapping relationship between wave dynamic characteristics and structural stress distribution, so as to achieve accurate and efficient analysis of stress distribution of permeable breakwaters under different wave conditions. Summary of the Invention

[0004] To address the aforementioned problems in existing technologies, the first aspect of this invention proposes a method for analyzing the stress distribution of permeable breakwaters under different wave conditions, comprising: S1: Based on sensors deployed in the structural parts and water-adjacent areas of the open breakwater, wave information and structural response information are collected in real time and synchronously to generate a synchronous wave-structure response dataset. S2: Based on the wave-structure response dataset, a wave-stress correlation map is constructed by analyzing the correspondence between wave characteristics and structural strain data; wherein, S2 includes: S201: Based on the wave-structure response dataset, group the data according to wave phase time series information and extract the strain data subsets corresponding to each phase time period; S202: Based on the strain data subsets corresponding to each phase time period, the data is processed by the map construction algorithm to generate a structural stress topology map that reflects the stress transfer relationship between the monitored parts. S203: Based on the wave phase time series information associated with the strain data subset and the structural stress topology map, establish the mapping relationship between phase information and topology map, and generate wave-stress correlation map. S3: Based on the characteristic information of the current wave condition to be analyzed and the wave-stress correlation spectrum, the correlation spectrum region corresponding to the current wave condition is determined through feature matching processing; wherein, S3 includes: S301: Extract features from the current wave condition and generate the key feature vector of the current wave condition; S302: Calculate the similarity between the key feature vectors of the current wave condition and the pre-stored wave feature vectors in the wave-stress correlation map; S303: Based on similarity, select the substructures of the graph that meet the preset similarity conditions to determine the associated graph regions; S4: Analyze the structural stress distribution information contained in the associated spectrum region to obtain the stress distribution of the permeable breakwater under the current wave conditions.

[0005] Compared with the prior art, the beneficial effects of the present invention are as follows: The technical solution of this invention effectively solves the problems raised in the background art through the close collaboration of four steps. First, step S1 generates a synchronized wave-structure response dataset by deploying sensors and collecting wave information and structural response information in real time, laying a data foundation and ensuring the accurate correspondence between wave dynamic characteristics and structural strain in the time dimension, overcoming the defects of traditional methods such as data asynchrony and inaccurate assumed loads. Next, step S2 constructs a wave-stress correlation map based on this dataset by analyzing the correspondence between wave characteristics and structural strain data; specifically, it extracts a subset of strain data by grouping according to wave phase time sequence information, constructs a structural stress topology map reflecting the stress transmission relationship between monitored parts, and then maps the phase information to the topology map to finally generate the map. This step transforms discrete time sequence data into a structured and visualized correlation model, establishes a direct mapping between wave dynamic phase and internal stress transmission path of the structure, and realizes a deep characterization of complex wave-structure interaction patterns. Then, step S3 determines the correlation map region corresponding to the current wave condition based on the feature information of the current wave condition to be analyzed and the constructed wave-stress correlation map through feature extraction, similarity calculation and matching. This step enables the rapid and accurate identification of the stress response mode that best matches the current wave conditions from a vast amount of historical correlation models, avoiding the need for complex simulation calculations to be performed again for each analysis, thus greatly improving the efficiency and relevance of the analysis. Finally, step S4 directly obtains the stress distribution of the permeable breakwater under the current wave conditions by analyzing the structural stress distribution information contained in the determined correlation map region.

[0006] The entire process forms a closed loop of "synchronous data acquisition—associative model construction—real-time wave condition matching—stress distribution analysis," with each step interconnected: S1 provides precisely synchronized input data; S2 establishes a universal wave-stress correlation knowledge base; S3 enables intelligent retrieval and matching for specific wave conditions; and S4 outputs the final analysis results. This method does not rely on artificial load assumptions but is directly driven by measured data, enabling it to dynamically and accurately reflect the true stress response of structures under different wave conditions, providing a reliable technical means for structural safety monitoring and assessment. Attached Figure Description

[0007] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0008] Figure 1 The diagram shown is a flowchart illustrating a method for analyzing the stress distribution of a permeable breakwater under different wave conditions, according to an embodiment of the present invention. Figure 2 The diagram shown is a structural schematic of a stress distribution analysis system for a permeable breakwater under different wave conditions, provided by an embodiment of the present invention. Detailed Implementation

[0009] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0010] The specific embodiments of the present invention will be described below.

[0011] Example 1 like Figure 1 As shown, the first aspect of this invention proposes a method for analyzing the stress distribution of a permeable breakwater under different wave conditions, comprising: S1: Based on sensors deployed in the structural parts and water-adjacent areas of the open breakwater, wave information and structural response information are collected in real time and synchronously to generate a synchronous wave-structure response dataset. S2: Based on the wave-structure response dataset, a wave-stress correlation map is constructed by analyzing the correspondence between wave characteristics and structural strain data; wherein, S2 includes: S201: Based on the wave-structure response dataset, group the data according to wave phase time series information and extract the strain data subsets corresponding to each phase time period; S202: Based on the strain data subsets corresponding to each phase time period, the data is processed by the map construction algorithm to generate a structural stress topology map that reflects the stress transfer relationship between the monitored parts. S203: Based on the wave phase time series information associated with the strain data subset and the structural stress topology map, establish the mapping relationship between phase information and topology map, and generate wave-stress correlation map. S3: Based on the characteristic information of the current wave condition to be analyzed and the wave-stress correlation spectrum, the correlation spectrum region corresponding to the current wave condition is determined through feature matching processing; wherein, S3 includes: S301: Extract features from the current wave condition and generate the key feature vector of the current wave condition; S302: Calculate the similarity between the key feature vectors of the current wave condition and the pre-stored wave feature vectors in the wave-stress correlation map; S303: Based on similarity, select the substructures of the graph that meet the preset similarity conditions to determine the associated graph regions; S4: Analyze the structural stress distribution information contained in the associated spectrum region to obtain the stress distribution of the permeable breakwater under the current wave conditions.

[0012] A method for stress distribution analysis of permeable breakwaters under different wave conditions begins with the systematic deployment of instruments for the target structure and its surrounding hydrodynamic environment. Permeable breakwaters typically consist of an upper panel, lower supporting columns, and possibly transverse and longitudinal connecting beams, forming an open structure. To capture its mechanical behavior, monitoring is required at "structural locations" that are sensitive to overall stress or prone to stress concentration. These locations are usually determined based on mechanical analysis or engineering experience, such as the mid-span area of ​​the panel, the connection nodes between the columns and the foundation or panel, and the midpoints or ends of the main beams. Strain sensors, such as resistance strain gauges or fiber optic grating sensors, are installed at these selected locations to detect the minute length changes, i.e., strain, produced by the structural material under load. This strain is directly related to the internal stress. Simultaneously, sensors for capturing wave information, such as wave height meters or wave radar, are deployed in the "waterfront area" in front of the breakwater—the water area where waves have not yet violently interacted with the structure. These sensors continuously record the sequence of wave surface elevation changes over time, implicitly containing the wave's "phase timing information," namely the order and duration of wave crests, troughs, rising edges, and falling edges on the time axis. By deploying a high-precision clock synchronization acquisition system, a unified "timestamp" is added to the "raw wave data" and "raw strain data" collected by all sensors at the moment of generation. This ensures that in subsequent processing, the wave state at any given time point can precisely correspond to the strain response generated by the structure at that time. This "real-time synchronous acquisition" and "data synchronous processing" process generates a strictly time-aligned "wave-structure response dataset," which is the data foundation of the entire method. It solves the problem of temporal disconnect between load input and structural response in traditional analysis, making it possible to establish accurate dynamic correlations.

[0013] Next, the core of the method is to uncover and solidify the inherent patterns in the data, namely, constructing a "wave-stress correlation map". This process is not simply data storage, but rather intelligent information purification and relationship modeling. First, based on the "phase time sequence information" in the wave data, continuous synchronous datasets are intelligently grouped. For example, each wave cycle can be identified and divided into characteristic "phase time periods", such as the peak period, trough period, and the rising period from trough to peak. Each such time period corresponds to a "strain data subset", which contains data recorded by all strain sensors within that time period. For each strain data subset, the relationship between strain signals at different monitoring points needs to be analyzed. This is achieved through a "map construction algorithm", which is usually based on graph theory principles, treating each monitoring location as a "node" in the graph. By calculating indices such as cross-correlation, coherence, or transfer function between strain data at different nodes, it is determined whether there is a strong "stress transfer relationship" between nodes and the strength of the relationship. This is used as the weight of the "edges", thereby drawing a series of "structural stress topology maps" reflecting how stress is transferred and distributed among points in the structural monitoring network under a specific wave phase. This diagram visually illustrates the path and key points of force flow. Finally, a clear mapping relationship is established between each structural stress topology diagram and the specific "wave phase timing information" upon which its generation depends (e.g., "Phase A: Crest Period"). When all phase time periods and their corresponding topology diagrams are linked, a "wave-stress correlation map" containing rich scenario knowledge is created. This map is essentially a structured database that systematically links dynamic wave processes with static stress transfer patterns, forming a queryable and accessible engineering knowledge base.

[0014] When faced with a new "current wave condition" requiring analysis, the method enters a rapid matching and parsing phase. The characteristic information of the current wave condition may originate from fragments extracted from real-time wave monitoring data or from a given set of design wave parameters. First, "feature extraction" is performed on this information to generate a digital summary that can be recognized and compared by machines, namely, the "key feature vector of the current wave condition." This process may include analyzing waveform features such as the principal period and asymmetry of the wave time history curve. Subsequently, the system calculates the "similarity" between this key feature vector and pre-stored feature vectors representing various historical wave states in the "wave-stress correlation map." Similarity calculation can employ metrics such as Euclidean distance and cosine similarity. Through comparison, the system can automatically filter out the "map substructures" corresponding to historical wave states that are most similar to the current wave characteristics from the vast map. The set of these selected substructures constitutes the "corresponding correlation map region" to the current wave condition. This process enables rapid and accurate location of the most relevant reference cases from a massive amount of historical patterns.

[0015] Ultimately, within the defined "correlation map region," "structural stress distribution information" corresponding to historical wave conditions highly similar to the current wave condition is stored. This information is directly reflected in the "structural stress topology map" contained within this region. By analyzing these topology maps, it becomes clear which parts of the structure are critical paths for stress transmission under wave conditions similar to the current one, which parts bear greater relative stress, and the stress correlation between different parts. Therefore, without complex real-time simulation calculations, the "stress distribution of the permeable breakwater under the current wave condition" can be directly derived.

[0016] The entire method creatively combines data-driven approaches with knowledge base retrieval through a logical chain of "synchronous data acquisition to establish precise data associations - constructing maps to solidify historical experience - feature matching for rapid retrieval - parsing maps to output distribution conclusions." Its technical advantage lies in its radical transformation of the traditional analysis model that relies on assumed loads and offline simulations. Synchronous data acquisition ensures the authenticity of causal relationships; constructing association maps transforms complex wave-structure interactions into queryable, structured knowledge, enabling the analysis process to learn and accumulate; feature matching achieves a second-level response to any new wave condition, resulting in extremely high analysis efficiency; and the final stress distribution conclusions are directly derived from models constructed from actual monitoring data, thus reflecting the actual service status of the structure more accurately than calculations based on traditional theoretical assumptions. This provides a dynamic, precise, and efficient technical means for the safety monitoring, condition assessment, and early warning of permeable breakwaters.

[0017] In some implementations, S1 includes: S101: Based on wave sensors deployed in the waterfront area, continuously collect raw wave data containing wave phase and timing information; S102: Based on strain sensors arranged in the structural parts, raw strain data is continuously collected; S103: By assigning a unified timestamp to the original wave data and the original strain data, data synchronization processing is performed to generate a synchronized wave-structure response dataset.

[0018] This embodiment further clarifies the specific implementation path of data acquisition, focusing on how to achieve the key prerequisite of "real-time synchronous acquisition" to ensure the quality of the basic data for all subsequent analyses. During implementation, an integrated physical monitoring system needs to be constructed. Wave sensors are deployed in the "water-adjacent area," such as pressure-type wave height meters, acoustic Doppler point current meters (which also have wave height measurement functions), or non-contact radar wave height meters. These sensors are fixed to underwater piles or floating platforms, and their core function is to continuously record the elevation changes of the water surface relative to the still water surface at the measuring point, thereby generating a series of "raw wave data" arranged in chronological order. This data not only contains the wave height amplitude, but more importantly, its precise time series, i.e., "wave phase time series information," which records the specific time and duration of each wave crest and trough.

[0019] Meanwhile, sensors are installed at the structural points of the permeable breakwater. Based on the structural design drawings and stress analysis, strain sensors are attached or welded to key stress points, such as the center and corners of the grating, the tops, middle and bases of the columns, and the mid-span and ends of the main and secondary beams. Resistance strain gauges are most commonly used, working on the principle that the resistance of a metal wire changes with its mechanical strain; fiber optic grating sensors can also be used, measuring strain through the wavelength of the grating. These sensors continuously convert minute deformations of the local structural materials into electrical or optical signals, recording them as "raw strain data." At this point, the two data streams from the water and the structure are physically acquired independently.

[0020] "Data synchronization processing" serves as the bridge connecting these two independent data streams, enabling them to interact and interact. Its core technology lies in "assigning a unified timestamp." This is not simply software time synchronization, but requires establishing a unified time reference at the hardware acquisition level. A typical implementation involves deploying a high-precision, high-stability clock source in the monitoring system, such as a GPS or BeiDou timing module. This clock source continuously broadcasts a precise absolute time signal to all wave sensor and strain sensor data acquisition units deployed in the field via a network (such as wired Ethernet or wireless LoRa) or a dedicated synchronization signal line. Each data acquisition unit, at the moment it records each data sample (such as a wave height or strain value), uses this broadcast time as a reference to assign a unique "timestamp" to the sample, consistent with the time references of all other acquisition units. After acquisition, the backend data processing center receives all the raw data with the unified timestamp. Subsequently, the processing software, based on this unified timeline, pairs, aligns, and packages wave data samples and strain data samples from the same moment or within the same small time window.

[0021] Through this series of operations, the previously separate streams of "raw wave data" and "raw strain data" are merged into a single, clearly defined, and time-corresponding "synchronized wave-structure response dataset." This step is crucial in that it fundamentally eliminates time asynchrony errors caused by internal clock drift, startup time differences, or network transmission delays in different devices. Such errors are fatal when analyzing the dynamic interaction between waves and structures, potentially leading to incorrect association of wave crest loads with wave trough responses. Precise synchronization ensures that every strain response in the dataset can be accurately attributed to its true wave excitation phase, allowing subsequent phase-based data grouping and correlation analysis to be built on a solid temporal logic foundation, greatly improving the reliability of the entire analysis method and the accuracy of the results.

[0022] In some implementations, in S101, the wave sensor is a wave height meter array, and S101 includes: S1011: Based on a wavefront altimeter array, it simultaneously acquires wavefront elevation data from multiple spatial points; S1012: Based on wavefront elevation data from multiple spatial points, three-dimensional wavefield information containing wave propagation direction and phase gradient is reconstructed through spatial interpolation and time series analysis. S1013: Integrates three-dimensional wave field information into the original wave data to generate original wave data containing three-dimensional wave field information.

[0023] This embodiment sets higher requirements for wave information acquisition methods, aiming to obtain spatial information about the wave field so that the analysis can more precisely consider the impact of wave directionality and spatial inhomogeneity on the structure. In this embodiment, the "wave sensors" deployed in the "waterfront area" are specifically "wave height meter arrays." This means that instead of using a single measuring point, multiple wave height meters with the same function are installed according to a pre-designed spatial layout. The array configuration needs to consider the orientation of the breakwater and the main wave direction. Common configurations include "linear arrays" perpendicular to the possible main wave direction, "grid arrays" covering a certain area, or "arc arrays" along the front edge of the breakwater. Each wave height meter occupies a known "spatial point" coordinate (obtained through measurement and positioning) in the array and independently and synchronously records the "wave surface elevation data" at that point. Therefore, the acquired raw data is a set of spatially discrete, temporally synchronized multi-point wave surface time series.

[0024] After obtaining multi-point synchronized data, the goal is not simply to store them side-by-side, but to "reconstruct" continuous wave field information through calculation. This process mainly relies on a combination of "spatial interpolation" and "time series analysis." "Spatial interpolation" is a mathematical method used to estimate the wave height at any location within the entire region of interest at the same time t, based on known values ​​at discrete spatial points (in this case, wave heights measured by various wave height meters at a certain time t). Commonly used interpolation algorithms include the inverse distance weighted method and the Kriging method. By performing such spatial interpolation at each sampling time, a series of spatial distribution maps describing the wave surface shape of the entire region at different instants can be obtained. Subsequently, "time series analysis" is performed on these instantaneous wave surface maps arranged in chronological order. By tracking the positional movement of specific characteristic points (such as the highest point of the wave crest) in the wave surface maps at adjacent times, the "propagation direction" of the wave can be calculated. By analyzing the lines formed by points with the same phase (such as wave crests) on the wave surface at the same time (i.e., the crest line), their direction and curvature can be observed. Simultaneously, the rate of change of wave phase (e.g., a periodic change from 0 to 360 degrees) in space, known as the "phase gradient," can be calculated. This reflects information about the wave's propagation speed and wavelength in space. The above processing ultimately "reconstructs three-dimensional wavefield information containing the wave propagation direction and phase gradient." Here, "three-dimensional" refers to two-dimensional horizontal space (x, y) plus wavefront elevation (z) or phase value.

[0025] Finally, the reconstructed three-dimensional wave field information is integrated into the original wave data as a whole attribute. This means that for each time stamp or set of timestamps, the corresponding wave data record not only contains the original wave height time series, but also includes derived spatial field information such as "the main wave direction at this moment is NNE" and "the phase gradient field distribution map at this moment." This information integration greatly enriches the descriptive dimensions of wave features. The resulting technical effects are profound: when constructing the "wave-stress correlation map" later, the raw materials used to generate the "wave feature vector" are no longer just statistical features of the time series, but can also include parameters such as direction and spatial phase difference. This allows the constructed map to distinguish between the different "structural stress topologies" induced by "waves from directly ahead" and "waves from an oblique direction," because oblique waves will generate phase difference loads at different spans of the structure. Similarly, when matching the "current wave condition," because the "key feature vector" contains spatial features such as wave direction, the accuracy and specificity of the matching are significantly improved, and the "correlation map region" that matches the current wave spatial characteristics can be located more accurately. Ultimately, based on the stress distribution obtained from this analysis, the stress response of a large-span spatial structure like a permeable breakwater under complex wave fields (such as oblique incident waves and short-peak irregular waves) can be more realistically reflected. This upgrades stress analysis from a rough "point-to-point" correlation to a refined "field-to-structure" mapping, significantly improving the engineering applicability and analytical accuracy of the method.

[0026] In some implementations, S301 includes: S3011: Based on the characteristic information of the current wave situation, identify the dominant wave phase pattern through waveform analysis; S3012: Based on the identified dominant wave phase pattern, the time-frequency domain distribution characteristics of the waves are extracted through spectral analysis; S3013: Based on the characteristics of the dominant wave phase mode and the time-frequency domain distribution characteristics, the key feature vector of the current wave condition is generated through vectorized combination.

[0027] This embodiment details the process of extracting digital features suitable for machine matching from the "feature information of the current wave condition," namely, the generation process of the "key feature vector," which is the core algorithmic step for achieving intelligent and accurate spectral matching. The input "feature information of the current wave condition" can take various forms: if based on real-time monitoring, it may be a newly acquired wave time history curve of finite length; if based on design or forecasting, it may be a set of parameters such as significant wave height, average period, spectral peak period, main wave direction, etc., or even spectral model parameters representing wave energy distribution (such as the γ parameter of the JONSWAP spectrum).

[0028] The first step is "waveform analysis," which aims to identify the "dominant wave phase pattern" from the macroscopic time morphology of waves. This is not simply reading parameters, but rather performing feature identification on the wave time-domain signal. For example, for a segment of wave time history data, continuous peaks and troughs can be identified by finding local maxima and minima, and then the stability of its periodic sequence, the distribution of wave height sequences, the symmetry of peaks and troughs (forward or backward tilt), and the presence of a significant long-period wave group envelope can be analyzed. Through this analysis, it can be determined whether the current wave is closer to a regular wave, an irregular wave with significant group characteristics, or a broken wave pattern. The identified "dominant wave phase pattern" is a high-level qualitative or semi-quantitative description of the wave time structure, which determines the basic temporal rhythm and impact characteristics of wave loads acting on the structure.

[0029] After grasping the macroscopic pattern, the second step is to conduct a deeper "spectral analysis" to extract the "time-frequency domain distribution characteristics" of the waves. The most classic spectral analysis method is to calculate the wave energy spectral density function, that is, to decompose the energy of the time-domain signal into different frequency components through Fourier transform. From the energy spectrum, a series of characteristic parameters can be extracted, such as the peak frequency (the frequency with the highest energy), the spectral width (the degree of concentration of energy distribution), and the spectral moments of each order (related to statistical quantities such as wave height and period). For non-stationary signals, "time-frequency analysis" tools, such as short-time Fourier transform or wavelet transform, are needed to observe how the energy of different frequency components evolves over time, thereby extracting the "joint time-frequency domain distribution characteristics," such as the pattern of energy change over time in a specific frequency band. This step reveals the complexity of the internal frequency composition of waves; two irregular waves with the same effective wave height and average period may produce different dynamic excitation effects on the structure due to the difference in their spectral shapes.

[0030] Finally, the results obtained from the above two steps are "vectorized and combined." This means representing the identified "dominant wave phase pattern" using a pre-defined set of codes or parameters (e.g., using a classification label [regular wave, wave group, ...] and a numerical value describing asymmetry). Simultaneously, the "time-frequency domain distribution characteristics" are quantified into multiple specific numerical parameters (such as spectral peak frequency fp, spectral width parameter ε, nth-order spectral moment mn, etc.). Then, according to a predefined, fixed order and format, these codes and numerical parameters are arranged into an ordered array, which is the "key feature vector of the current wave condition." This vector is a multi-dimensional, digital "fingerprint" that comprehensively and concisely represents all the key attributes of the current wave in both the time-domain morphology and frequency-domain structure. Its technical advantage lies in providing a standardized and computable feature extraction process, capable of transforming wave condition information of any input form into a machine-readable and comparable unified format. By combining waveform and spectrum analysis, the generated feature vector contains both macroscopic pattern information of wave action and microscopic spectral details, exhibiting strong characterization and discriminative power. When this high-quality feature vector is compared with historical feature vectors in step S302 for "similarity" calculation, it achieves precise matching based on the intrinsic dynamic characteristics of waves, rather than relying solely on a few macroscopic statistical parameters. This ensures that the system can select historical records from the "wave-stress correlation map" that are most similar to the current situation in terms of the essential characteristics of waves, thus providing the most relevant and reliable data support for subsequent extraction and analysis of stress distribution information. This is the key algorithmic guarantee for the entire method to achieve accurate and adaptive analysis.

[0031] In some implementations, S4 includes: S401: Based on the correlation map region, extract the corresponding structural stress topology map and wave load data from the wave-stress correlation map containing load information; S402: Based on wave load data, a localized refined mechanical simulation calculation of the open breakwater structure is performed using finite element analysis software to obtain localized refined stress results; S403: Based on the structural stress topology diagram and local refined stress results, the stress distribution of the permeable breakwater under the current wave conditions is obtained through data fusion and calibration.

[0032] This implementation aims to specifically illustrate how to extract sufficiently accurate stress distribution data for engineering judgment from matched correlation map regions. Its core lies in constructing a two-stage analysis process: first, extracting macroscopic, correlated guiding information from the map; then, driving a targeted, localized, refined mechanical calculation based on this information; and finally, integrating the results of both. Once the system determines the correlation map region corresponding to the current wave condition through feature matching, the information stored within this region is not a single data point, but a correlation information package. The "analysis" operation in step S401 extracts two core components from this information package. The first is a "structural stress topology map," a graph-based model that abstracts monitoring points as "nodes" and the mutual influence and transmission relationships of stress between points as weighted "edges." This visually displays how stress is transmitted from the loading point to other parts through the structural network under specific wave conditions, and identifies key force transmission paths and stress concentration nodes. The second is "wave load data," which is a mechanical input directly related to the historical wave conditions used to generate the topology map. This data can take the form of pressure time histories or concentrated force time histories acting on structural feature locations (such as the center of a panel or the wave-facing surface of a column), or statistically analyzed characteristic forces (such as maximum impact force or average pressure). Extracting these two data points provides clear load inputs and guidance for key areas of focus in the next stage of refined simulation.

[0033] Next, step S402 initiates a more precise mechanical analysis process: "performing localized refined mechanical simulation calculations of the permeable breakwater structure using finite element analysis software." Finite element analysis is a widely used numerical computation method that discretizes a continuous physical structure into a finite number of small elements (such as tetrahedral or hexahedral elements), connected by nodes. By solving a large set of equations based on physical laws (such as equilibrium equations and constitutive relations), the response of the structure under load is simulated. Here, "localized refinement" is the key feature of this step. It means that instead of performing high-precision meshing and calculations across the entire massive breakwater structure—which would be computationally intensive—a highly detailed finite element model is created only for the local structural regions involved in these critical paths (e.g., certain connected panels and their underlying supporting columns). This model captures detailed stress variations. Simultaneously, the extracted wave load data is applied to this locally refined model in a form that conforms to the actual physical distribution. After running the calculation, the resulting "locally refined stress results" will provide accurate stress values, gradient distributions, and possible stress concentration points within these key areas, with an accuracy far exceeding that of relative judgments based on topological relationships.

[0034] However, high-precision local results lack coordination with the overall structure. Therefore, the "data fusion and calibration processing" in step S403 is crucial. Fusion refers to combining macroscopic topological relationships with precise microscopic values. Calibration is a standardization process. Specifically, the "local refined stress results" can be used as a benchmark. For example, refined calculations show that the stress value of point A, marked as a key node in the topology diagram, is σ_A. Then, in the "structural stress topology diagram," the stress relationship of point B, connected to point A by an "edge," relative to point A (derived from the edge weight or topological algorithm, for example, the stress at point B is approximately 0.7 times that at point A), can be calibrated to calibrate the estimated stress value at point B to approximately 0.7 × σ_A. Simultaneously, if the local refined model also includes point B or its surrounding area, the calculation result at that point can be used as verification to fine-tune the aforementioned topology-based extrapolation values. By "injecting" precise values ​​into the topological relationship network, the topology diagram, which originally only showed relative relationships, is given a real stress scale. Meanwhile, for other structural parts that are not directly covered by the refined model but are connected to the key areas through topological relationships, reasonable stress estimation can also be performed based on the calibrated topology map.

[0035] Ultimately, through this fusion and calibration process, a complete picture of the stress distribution of a permeable breakwater under the current wave conditions is generated. This picture reflects both the overall stress transmission pattern (embodied in the topology) guaranteed by the correlation of historical data and has been accurately verified and calibrated mechanically. The technical advantage of this method lies in its ingenious solution to the problem of excessively high costs associated with comprehensive high-precision simulation calculations, while pure data-driven topology analysis lacks absolute mechanical accuracy. By intelligently guiding "where fine-grained calculations are needed" through correlation maps, it achieves optimized allocation of computing resources. Furthermore, through fusion calibration, the results of local high-precision "points" are reliably extended to the "surface" represented by the topology map. Thus, at an acceptable computational cost, a comprehensive picture of stress distribution with both overall reliability and local accuracy is obtained, providing high-quality and efficient data support for engineering safety assessment.

[0036] In some implementations, after S4, the following is also included: S5: Based on the stress distribution of the permeable breakwater under the current wave conditions, identify key structural areas with stress exceeding the preset level by comparing thresholds; S6: Based on the stress distribution characteristics of the key structural regions, a structural parameter adjustment scheme is generated through a structural parameter optimization algorithm.

[0037] This implementation extends the endpoint of the method from "state analysis" to "decision support," adding a step of proactively identifying risks and generating response suggestions. After obtaining the stress distribution of the permeable breakwater under the current wave conditions through the aforementioned steps, the system first executes step S5: identifying key structural areas of concern where stress exceeds a preset level through threshold comparison. Here, "stress distribution" includes the stress values ​​of each analyzed part of the structure. The "preset level" is a crucial safety judgment benchmark; it is not arbitrarily set but is a permissible stress value determined comprehensively based on the material properties used in the structural design (such as the design tensile strength of concrete and the yield strength of steel), the importance level of the structure, and the safety factor required by relevant engineering specifications. This preset level may differ for different parts or different materials. The system automatically compares the calculated stress value at each location in the stress distribution with the corresponding preset level. Once the calculated stress value at a certain location exceeds the preset level set for it, that location and its adjacent affected area are automatically marked and recorded by the system as "key structural areas of concern." This process enables automatic screening and risk point location of massive stress data, quickly and objectively highlighting potentially hazardous areas from the overall structure, avoiding the omissions and inefficiencies that may occur with manual point-by-point analysis.

[0038] After identifying the risk areas, the method does not stop there, but proceeds to a more constructive step, S6: based on the stress distribution characteristics of these areas, a structural parameter adjustment scheme is generated through a structural parameter optimization algorithm. "Stress distribution characteristics" is a detailed description of the stress state of the key areas of concern, including not only the specific values ​​of the exceeding stresses, but also their distribution pattern within the area (e.g., uniform distribution, gradient distribution, local spikes), their variation over time (corresponding to wave phases), and their correlation and coupling relationship with the stresses of adjacent areas. These characteristics collectively constitute a mechanical profile of why the stress in this area exceeds the limit. "Structural parameters," on the other hand, refer to physical or geometric variables that can be adjusted artificially during structural design or modification, such as the thickness of a reinforced concrete panel, the outer diameter and wall thickness of a steel pipe column, the steel plate size or weld type of a connection node, and the spacing and height of local reinforcing ribs. "Structural parameter optimization algorithms" are a class of mathematical optimization algorithms, such as genetic algorithms, simulated annealing algorithms, or sequential quadratic programming algorithms. They can find the optimal solution in a multidimensional solution space composed of multiple adjustable structural parameters according to certain search strategies (such as simulating biological evolution, physical annealing processes, or gradient-based methods). In this scenario, the standard for "optimization" is very clear: it means that under the same current wave condition characteristics, the stress level of the adjusted structure in these key areas of concern can be reduced until the preset level requirement is met.

[0039] The technical advantage of this implementation lies in its completion of a full logical loop from "diagnosis" to "solution." Through automated threshold comparison, the method achieves rapid and quantitative assessment of structural safety status and accurately identifies weak points. More importantly, it doesn't stop at simply identifying problems; it further utilizes optimization algorithms to transform specific engineering problems (such as excessive stress at a certain location) into mathematical parameter optimization problems. By simulating various possible structural parameter adjustment schemes (in a virtual environment) and evaluating their effectiveness in improving the stress state of the target area, the algorithm can ultimately recommend one or more effective "structural parameter adjustment schemes." This is equivalent to providing engineers with data-driven, targeted reinforcement or design optimization suggestions, such as "it is recommended to increase the panel thickness in area A by X% and add a reinforcing plate at node B." This makes the method not only an analysis tool but also an intelligent system that assists in design decision-making, greatly enhancing its practical value in the design review of permeable breakwaters, the improvement of existing structural safety, and the formulation of renovation plans.

[0040] In some implementations, S6 includes: S601: A structural stress prediction model is obtained by training a machine learning algorithm based on historical wave-structure response data. S602: Input the stress distribution characteristics and current wave condition features of the key structural regions into the structural stress prediction model. S603: Based on the initial output of the structural stress prediction model, with the goal of optimizing the stress level of the key structural region, the structural design parameters in the input of the structural stress prediction model are iteratively adjusted so that the prediction output of the structural stress prediction model meets the optimization conditions. S604: Based on the set of structural design parameters used in the structural stress prediction model when the optimization conditions are met, generate a structural parameter adjustment scheme.

[0041] This implementation provides a specific technical means for efficiently generating structural parameter adjustment schemes. Its core idea is to utilize a machine learning model as a proxy for rapid stress prediction, combined with an optimization algorithm, to replace the traditional time-consuming "simulation-evaluation" trial-and-error cycle. This approach begins with an offline model building phase, namely step S601: based on historical wave-structural response data, a structural stress prediction model is trained using a machine learning algorithm. Here, the "historical wave-structural response data" is a rich training sample set. Each sample contains two parts: first, "input features," typically composed of wave feature vectors (such as wave height, period, and directional spectrum parameters) and structural design parameter vectors (such as the dimensions and material properties of each component); second, "output labels," which are the "stress distribution characteristics of the key structural regions" corresponding to the input conditions, obtained through reliable methods (such as high-precision finite element simulation or inversion from actual long-term monitoring data). These characteristics may be quantified into a set of feature parameters such as maximum stress value, average stress value, and stress non-uniformity coefficient. The selected "machine learning algorithm" can be a model capable of handling complex nonlinear relationships, such as deep neural networks, gradient boosting decision trees, or support vector regression. The training process involves using a large number of such sample pairs to allow the algorithm to automatically learn the complex mapping function between "input features" and "output stress characteristics." After training is complete and the prediction accuracy is ensured through validation set testing, the resulting "structural stress prediction model" has the ability to predict stress response based on given wave and structural parameters within "seconds" or even "milliseconds," which is several orders of magnitude faster than running a finite element analysis.

[0042] During the online application phase, when optimization is needed for identified key structural areas of interest, step S602 sends the necessary input information to the prediction model. The input information mainly includes two parts: first, the "current wave condition characteristics" that induce the current stress state, which have been extracted as feature vectors; and second, the initial state of the "structural design parameters" for the area, i.e., the initial values ​​of the set of parameters to be optimized. After receiving these inputs, the model quickly performs forward propagation calculations and outputs an "initial predicted stress distribution," which is essentially a rapid reproduction of the expected stress level of the area under the current unoptimized state, providing a baseline for subsequent optimization.

[0043] Subsequently, step S603 enters the core iterative optimization loop. The objective function of the optimization is clear and quantifiable: to optimize (reduce) the stress level in the key structural region. The optimization operation targets the "structural design parameters" in the input model. The system's built-in "strategy optimization algorithm" (such as genetic algorithm, particle swarm optimization, etc.) begins to work. It first generates a set of adjustment suggestions for the structural design parameters based on the problem reflected by the "initial predicted stress distribution," that is, it generates "new structural parameters." Then, step S603 takes the unchanged "current wave condition characteristics" and this set of "new structural parameters" as a new input combination and inputs it into the "structural stress prediction model" again. The model makes a rapid prediction based on the new parameters to obtain the "updated predicted stress distribution." The system evaluates the result of this prediction to determine whether the stress in the key region meets the "optimization conditions" (e.g., whether the maximum stress value is lower than the preset level). If not, the process of steps S6032 and S6033 is repeated: the optimization algorithm adjusts the parameters again according to the latest prediction results, generates the next set of "new structural parameters," and then calls the model again for prediction evaluation. This cycle of "adjusting parameters - model prediction - result evaluation" is repeated repeatedly.

[0044] Finally, after several iterations, a set of parameters will eventually emerge that causes the stress in the key structural region to reach the preset optimization target in the model's predicted output, i.e., "the predicted output of the structural stress prediction model satisfies the optimization conditions." At this point, the iteration stops. Step S604 is then executed: the system records the specific set of "structural design parameters" used in the last prediction that satisfies the conditions. This set of parameter values ​​represents the solution found throughout the optimization process that effectively improves the stress state of the target region. Based on this parameter set, a specific "structural parameter adjustment scheme" can be generated, specifying which dimensions of which components need to be adjusted to what values. The technical effect of this implementation method is extremely significant. By introducing a machine learning model as a high-speed, high-fidelity surrogate model, it completely changes the predicament of traditional structural optimization relying on repetitive and time-consuming simulations. The optimization algorithm can explore thousands of different parameter combinations in a short time, and the surrogate model provides real-time performance predictions, making global optimization possible in a vast design space. This greatly improves the efficiency and depth of optimization design, helping engineers discover better parameter combinations that are difficult to conceive of based on experience, thus providing a powerful technical tool for achieving refined and intelligent performance enhancement of complex structures such as permeable breakwaters.

[0045] In some implementations, S603 includes: S6031: Input the current wave condition characteristics and initial structural parameters into the structural stress prediction model to obtain the initial predicted stress distribution; S6032: Based on the initial predicted stress distribution, with the goal of reducing stress in the key structural regions, the structural parameters are adjusted through a strategy optimization algorithm to generate new structural parameters; S6033: Input the current wave condition characteristics and new structural parameters into the structural stress prediction model to obtain the updated predicted stress distribution; S6034: Repeat the adjustment and prediction process of S6032 and S6033 until the stress in the key structural region of the obtained predicted stress distribution is lower than the preset level. At this time, the prediction output of the structural stress prediction model meets the optimization conditions.

[0046] This implementation provides a more detailed and progressive breakdown of the aforementioned iterative optimization loop, clearly depicting the entire process of gradually approaching the optimal solution from the initial state through a "prediction-adjustment-reprediction" feedback mechanism, ensuring the robustness and operability of the optimization process. The process begins with a clear initial state assessment, namely step S6031: inputting the "current wave condition characteristics" describing the external environment and the "initial structural parameters" (i.e., original design parameters or current state parameters) describing the structure's own state into a well-trained "structural stress prediction model." The model performs internal calculations based on this input and outputs the corresponding "initial predicted stress distribution." This distribution map quantitatively demonstrates the theoretical stress state of the structure under the target wave conditions before any optimization improvements, particularly clearly indicating the predicted stress level of the "key structural areas" under the initial design. This establishes a benchmark and starting point for comparison throughout the optimization process.

[0047] After obtaining the initial baseline, the optimization iteration loop officially begins. Step S6032 sets a clear and singular optimization direction: reducing the stress in the "key structural regions" is the primary and quantifiable objective. To achieve this objective, the system invokes the embedded "strategy optimization algorithm." This algorithm analyzes the information revealed by the "initial predicted stress distribution" (e.g., which locations have the highest stress and by how much it exceeds limits) and generates a first set of modification suggestions based on its specific search logic. If a genetic algorithm is used, it may "generate new structural parameters" by simulating "selection" (retaining parameter features corresponding to lower stress), "crossover" (mixing different parameter combinations), and "mutation" (randomly changing certain parameters). If gradient descent is used, the gradient of the stress objective function with respect to the structural parameters is calculated, and the parameters are adjusted along the negative gradient direction. This newly generated set of parameters is a specific modification proposal for the original design.

[0048] The new parameter combination must be validated. Therefore, step S6033 immediately follows with validation: the system combines the unchanged "current wave condition characteristics" with the "new structural parameters" just generated by the algorithm to form a completely new set of input conditions, which is then submitted to the "structural stress prediction model". The model performs rapid forward calculations based on this new set of inputs and outputs the "updated predicted stress distribution". The result of this prediction simulates the expected stress state if the structure is modified according to the parameter suggestions proposed in the previous step.

[0049] Next comes the crucial convergence assessment step. The system compares the "updated predicted stress distribution" with the preset optimization target, focusing on whether the stress in the "key structural region" is now "below the preset level." If the result is negative, meaning the stress is still higher than or equal to the preset level, it indicates that the current parameter adjustment has not achieved the expected effect, and optimization needs to be further deepened. At this point, the process explicitly instructs to "repeat the adjustment and prediction processes of S6032 and S6033." This means that the system will use the latest "updated predicted stress distribution" as the new analysis basis and starting point, replacing the initial "preliminary predicted stress distribution." Then, step S6032 is restarted: the strategy optimization algorithm, based on this new stress distribution that reflects the effect of the previous round of adjustments, re-analyzes and calculates, generating the next round (potentially more refined and more accurate) of "new structural parameters." Next, step S6033 is executed again: the new parameter combination is used for the next round of model prediction to obtain the updated stress distribution. This closed loop of "analyzing the problem based on the latest results - generating adjustment schemes - virtually verifying the effectiveness of the schemes" is repeatedly executed.

[0050] This iterative process continues, much like an automated process of trial and error and approximation in a virtual design space. Each iteration improves upon the results of the previous one. Until, after a certain iteration, the "updated predicted stress distribution" obtained in step S6033 shows that the stress of the continuously monitored structures has finally successfully decreased to within a safe range, that is, truly "below the preset level." At this point, the termination condition of the loop is triggered because "the predicted output of the structural stress prediction model meets the optimization conditions." The technical advantage of this implementation is that it defines a rigorous, closed-loop, and automatically executable digital optimization workflow. By explicitly starting with the initial state prediction, using "analyzing the current stress - intelligently adjusting parameters - rapidly predicting the effect" as the basic loop unit, and using the satisfaction of engineering safety standards as a hard termination condition, the entire process is logically rigorous, with clear entry, execution, and exit mechanisms. This step-by-step iterative method can effectively handle the complex, nonlinear, and even non-monotonic relationships between structural response and multiple design parameters, gradually approaching the optimal solution region through rapid virtual experiments supported by machine learning models. It ensures that the final "structural parameter adjustment scheme" is not a blind guess, but a reliable solution obtained through multiple virtual simulation verifications and gradual convergence. This significantly improves the scientific nature and engineering credibility of the optimization scheme, laying a solid operational foundation for the automated and intelligent performance optimization of complex hydraulic structures.

[0051] Example 2 like Figure 2 As shown, in a second aspect, the present invention proposes a stress distribution analysis system for permeable breakwaters under different wave conditions. The system employs a stress distribution analysis method for permeable breakwaters under different wave conditions proposed in any of the above embodiments. The system includes: The data synchronization acquisition module is used to perform step S1: based on the sensors arranged in the structural parts and water-adjacent areas of the open breakwater, it synchronously acquires wave information and structural response information in real time and generates a synchronized wave-structure response dataset. The correlation map construction module is used to execute step S2: based on the wave-structure response dataset, a wave-stress correlation map is constructed by analyzing the correspondence between wave characteristics and structural strain data; wherein, S2 includes: S201: Based on the wave-structure response dataset, group the data according to wave phase time series information and extract the strain data subsets corresponding to each phase time period; S202: Based on the strain data subsets corresponding to each phase time period, the data is processed by the map construction algorithm to generate a structural stress topology map that reflects the stress transfer relationship between the monitored parts. S203: Based on the wave phase time series information associated with the strain data subset and the structural stress topology map, establish the mapping relationship between phase information and topology map, and generate wave-stress correlation map. The map matching and localization module is used to execute step S3: based on the feature information of the current wave condition to be analyzed and the wave-stress correlation map, the correlation map region corresponding to the current wave condition is determined through feature matching processing; wherein, S3 includes: S301: Extract features from the current wave condition and generate the key feature vector of the current wave condition; S302: Calculate the similarity between the key feature vectors of the current wave condition and the pre-stored wave feature vectors in the wave-stress correlation map; S303: Based on similarity, select the substructures of the graph that meet the preset similarity conditions to determine the associated graph regions; The stress distribution analysis module is used to execute step S4: analyze the structural stress distribution information contained in the associated spectrum region to obtain the stress distribution of the permeable breakwater under the current wave conditions.

[0052] This system corresponds to the method proposed in Example 1, and will not be described in detail here.

[0053] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for analyzing the stress distribution of a permeable breakwater under different wave conditions, characterized in that, include: S1: Based on sensors deployed in the structural parts and water-adjacent areas of the open breakwater, wave information and structural response information are collected in real time and synchronously to generate a synchronous wave-structure response dataset. S2: Based on the wave-structure response dataset, a wave-stress correlation map is constructed by analyzing the correspondence between wave characteristics and structural strain data; wherein, S2 includes: S201: Based on the wave-structure response dataset, group the data according to wave phase time series information and extract the strain data subsets corresponding to each phase time period; S202: Based on the strain data subsets corresponding to each phase time period, the data is processed by the map construction algorithm to generate a structural stress topology map that reflects the stress transfer relationship between the monitored parts. S203: Based on the wave phase time series information associated with the strain data subset and the structural stress topology map, establish the mapping relationship between phase information and topology map, and generate wave-stress correlation map. S3: Based on the characteristic information of the current wave condition to be analyzed and the wave-stress correlation spectrum, the correlation spectrum region corresponding to the current wave condition is determined through feature matching processing; wherein, S3 includes: S301: Extract features from the current wave condition and generate the key feature vector of the current wave condition; S302: Calculate the similarity between the key feature vectors of the current wave condition and the pre-stored wave feature vectors in the wave-stress correlation map; S303: Based on similarity, select the substructures of the graph that meet the preset similarity conditions to determine the associated graph regions; S4: Analyze the structural stress distribution information contained in the associated spectrum region to obtain the stress distribution of the permeable breakwater under the current wave conditions.

2. The method for stress distribution analysis of a permeable breakwater under different wave conditions according to claim 1, characterized in that, S1 includes: S101: Based on wave sensors deployed in the waterfront area, continuously collect raw wave data containing wave phase and timing information; S102: Based on strain sensors arranged in the structural parts, raw strain data is continuously collected; S103: By assigning a unified timestamp to the original wave data and the original strain data, data synchronization processing is performed to generate a synchronized wave-structure response dataset.

3. The method for stress distribution analysis of a permeable breakwater under different wave conditions according to claim 2, characterized in that, In S101, the wave sensor is a wave height meter array, and S101 includes: S1011: Based on a wavefront altimeter array, it simultaneously acquires wavefront elevation data from multiple spatial points; S1012: Based on wavefront elevation data from multiple spatial points, three-dimensional wavefield information containing wave propagation direction and phase gradient is reconstructed through spatial interpolation and time series analysis. S1013: Integrates three-dimensional wave field information into the original wave data to generate original wave data containing three-dimensional wave field information.

4. The method for stress distribution analysis of a permeable breakwater under different wave conditions according to claim 1, characterized in that, S301 includes: S3011: Based on the characteristic information of the current wave situation, identify the dominant wave phase pattern through waveform analysis; S3012: Based on the identified dominant wave phase pattern, the time-frequency domain distribution characteristics of the waves are extracted through spectral analysis; S3013: Based on the characteristics of the dominant wave phase mode and the time-frequency domain distribution characteristics, the key feature vector of the current wave condition is generated through vectorized combination.

5. The method for stress distribution analysis of a permeable breakwater under different wave conditions according to claim 1, characterized in that, S4 include: S401: Based on the correlation map region, extract the corresponding structural stress topology map and wave load data from the wave-stress correlation map containing load information; S402: Based on wave load data, a localized refined mechanical simulation calculation of the open breakwater structure is performed using finite element analysis software to obtain localized refined stress results; S403: Based on the structural stress topology diagram and local refined stress results, the stress distribution of the permeable breakwater under the current wave conditions is obtained through data fusion and calibration.

6. The method for stress distribution analysis of a permeable breakwater under different wave conditions according to claim 1, characterized in that, Following S4, it also includes: S5: Based on the stress distribution of the permeable breakwater under the current wave conditions, identify key structural areas with stress exceeding the preset level by comparing thresholds; S6: Based on the stress distribution characteristics of the key structural regions, a structural parameter adjustment scheme is generated through a structural parameter optimization algorithm.

7. The method for stress distribution analysis of a permeable breakwater under different wave conditions according to claim 6, characterized in that, S6 include: S601: A structural stress prediction model is obtained by training a machine learning algorithm based on historical wave-structure response data. S602: Input the stress distribution characteristics and current wave condition features of the key structural regions into the structural stress prediction model. S603: Based on the initial output of the structural stress prediction model, with the goal of optimizing the stress level of the key structural region, the structural design parameters in the input of the structural stress prediction model are iteratively adjusted so that the prediction output of the structural stress prediction model meets the optimization conditions. S604: Based on the set of structural design parameters used in the structural stress prediction model when the optimization conditions are met, generate a structural parameter adjustment scheme.

8. The method for stress distribution analysis of a permeable breakwater under different wave conditions according to claim 7, characterized in that, S603 includes: S6031: Input the current wave condition characteristics and initial structural parameters into the structural stress prediction model to obtain the initial predicted stress distribution; S6032: Based on the initial predicted stress distribution, with the goal of reducing stress in the key structural regions, the structural parameters are adjusted through a strategy optimization algorithm to generate new structural parameters; S6033: Input the current wave condition characteristics and new structural parameters into the structural stress prediction model to obtain the updated predicted stress distribution; S6034: Repeat the adjustment and prediction process of S6032 and S6033 until the stress in the key structural region of the obtained predicted stress distribution is lower than the preset level. At this time, the prediction output of the structural stress prediction model meets the optimization conditions.

9. A stress distribution analysis system for a permeable breakwater under different wave conditions, characterized in that, The system employs a stress distribution analysis method for permeable breakwaters under different wave conditions as described in any one of claims 1 to 8. The system comprises: The data synchronization acquisition module is used to perform step S1: based on the sensors arranged in the structural parts and water-adjacent areas of the open breakwater, it synchronously acquires wave information and structural response information in real time and generates a synchronized wave-structure response dataset. The correlation map construction module is used to execute step S2: based on the wave-structure response dataset, a wave-stress correlation map is constructed by analyzing the correspondence between wave characteristics and structural strain data; wherein, S2 includes: S201: Based on the wave-structure response dataset, group the data according to wave phase time series information and extract the strain data subsets corresponding to each phase time period; S202: Based on the strain data subsets corresponding to each phase time period, the data is processed by the map construction algorithm to generate a structural stress topology map that reflects the stress transfer relationship between the monitored parts. S203: Based on the wave phase time series information associated with the strain data subset and the structural stress topology map, establish the mapping relationship between phase information and topology map, and generate wave-stress correlation map. The map matching and localization module is used to execute step S3: based on the feature information of the current wave condition to be analyzed and the wave-stress correlation map, the correlation map region corresponding to the current wave condition is determined through feature matching processing; wherein, S3 includes: S301: Extract features from the current wave condition and generate the key feature vector of the current wave condition; S302: Calculate the similarity between the key feature vectors of the current wave condition and the pre-stored wave feature vectors in the wave-stress correlation map; S303: Based on similarity, select the substructures of the graph that meet the preset similarity conditions to determine the associated graph regions; The stress distribution analysis module is used to execute step S4: analyze the structural stress distribution information contained in the associated spectrum region to obtain the stress distribution of the permeable breakwater under the current wave conditions.

Images