A real-time calculation method and system for wave force of a breakwater

Abstract

The present application provides a kind of breakwater wave force real-time calculation method and system, it is related to port safety monitoring technical field.The method is by wave monitoring sensor array and structural response monitoring sensor network synchronous acquisition wave parameter and structural response data;Based on wave parameter and initial wave force calculation model, obtain initial wave force value;Utilize structural response data through structural mechanics model conversion and numerical difference analysis, obtain error signal;Combined with error signal and environmental state monitoring data, using reinforcement learning algorithm dynamically adjusts model parameter;Finally, based on the model parameter of optimization, recalculate to obtain real-time wave force value.The present application realizes the dynamic correction and self-adaptive optimization of wave force calculation, effectively improves the accuracy and real-time of breakwater safety monitoring.

Description

Technical Field

[0001] This invention relates to the field of port safety monitoring technology, and in particular to a method and system for real-time calculation of wave force on breakwaters. Background Technology

[0002] Traditional single-point prevention and control models primarily target individual risk sources, such as installing surveillance cameras or increasing personnel patrols. However, when faced with coupled multi-source risks (such as equipment failure combined with severe weather), they often fail to achieve full coverage, resulting in blind spots in safety management. This is especially true for critical infrastructure such as breakwaters, whose stability directly impacts the overall safety of the port. Traditional wave force calculation methods largely rely on static models or offline data, which cannot adapt to dynamic environmental changes in real time. This leads to inaccurate risk assessments, delayed responses, and potentially serious consequences such as structural instability.

[0003] Therefore, there is an urgent need for a prevention and control technology that can calculate wave forces in real time and accurately and dynamically optimize models to improve the inherent safety level of ports and ensure the resilience of the supply chain. Summary of the Invention

[0004] To address the aforementioned problems in the existing technology, the first aspect of this invention proposes a method for real-time calculation of wave force on breakwaters, comprising:

[0005] S1: Based on the wave monitoring sensor array and structural response monitoring sensor network installed on the breakwater, wave parameters including wave height and wave incident angle and structural response data including structural strain and vibration response are obtained by synchronously collecting wave physical characteristic data and structural dynamic response data.

[0006] S2: Based on wave parameters and the initial wave force calculation model pre-stored in the calculation unit, the initial wave force value is obtained by calling the basic parameters of the model and executing the wave force calculation formula.

[0007] S3: Based on structural response data and initial wave force values, an error signal reflecting the magnitude of the calculation error is obtained through structural mechanics model transformation and numerical difference analysis.

[0008] S4: Based on error signals and real-time environmental condition monitoring data, the model parameters are dynamically adjusted through the strategy optimization module of the reinforcement learning algorithm to obtain the optimized wave force calculation model parameters;

[0009] S5: Based on the optimized wave force calculation model parameters, the real-time wave force value is obtained by updating the calculation model parameter configuration and re-executing the wave force calculation.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0011] This approach effectively solves the problems of traditional single-point control models in the background technology, such as their inability to adapt to dynamic environments in real time and inaccurate risk assessments. First, based on a wave monitoring sensor array and a structural response monitoring sensor network installed on the breakwater, wave physical characteristic data and structural dynamic response data are collected synchronously. This yields wave parameters including wave height and wave incidence angle, and structural response data including structural strain and vibration response. This step enables multi-dimensional real-time monitoring of the breakwater's condition, providing a comprehensive and synchronous data foundation for subsequent calculations and overcoming the shortcomings of traditional methods, such as single data sources and delayed updates. Next, based on the wave parameters and the initial wave force calculation model pre-stored in the calculation unit, the initial wave force value is obtained by calling the model's basic parameters and executing the wave force calculation formula. This step utilizes the preset model to quickly generate preliminary assessment results, providing fundamental support for real-time response. Then, based on the structural response data and the initial wave force value, error signals reflecting the magnitude of calculation errors are obtained through structural mechanics model transformation and numerical difference analysis. This step accurately identifies calculation deviations by comparing measured data with model output, providing crucial basis for model optimization. Furthermore, based on error signals and real-time environmental monitoring data, the model parameters are dynamically adjusted using the strategy optimization module of a reinforcement learning algorithm to obtain optimized wave force calculation model parameters. This step leverages the adaptive capability of reinforcement learning to continuously adjust the model according to environmental changes and error feedback, significantly improving calculation accuracy and dynamic adaptability. Finally, based on the optimized wave force calculation model parameters, the wave force calculation is re-executed by updating the calculation model parameter configuration, resulting in real-time wave force values. This achieves real-time updating of model parameters and accurate output of wave force.

[0012] Overall, these steps, through a closed-loop collaboration of data acquisition, model calculation, error analysis, parameter optimization, and real-time output, achieve dynamic accuracy and adaptive adjustment in wave force calculation, effectively addressing the problems of inaccurate risk prediction and lagging prevention and control in complex environments, and enhancing the initiative and reliability of port safety and control. Attached Figure Description

[0013] 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.

[0014] Figure 1 The diagram shown is a flowchart illustrating a real-time wave force calculation method for breakwaters according to an embodiment of the present invention.

[0015] Figure 2The diagram shown is a structural schematic of a real-time wave force calculation system for a breakwater provided in an embodiment of the present invention. Detailed Implementation

[0016] 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.

[0017] The specific embodiments of the present invention will be described below.

[0018] Example 1

[0019] like Figure 1 As shown, this invention proposes a real-time calculation method for wave forces on breakwaters, including:

[0020] S1: Based on the wave monitoring sensor array and structural response monitoring sensor network installed on the breakwater, wave parameters including wave height and wave incident angle and structural response data including structural strain and vibration response are obtained by synchronously collecting wave physical characteristic data and structural dynamic response data.

[0021] S2: Based on wave parameters and the initial wave force calculation model pre-stored in the calculation unit, the initial wave force value is obtained by calling the basic parameters of the model and executing the wave force calculation formula.

[0022] S3: Based on structural response data and initial wave force values, an error signal reflecting the magnitude of the calculation error is obtained through structural mechanics model transformation and numerical difference analysis.

[0023] S4: Based on error signals and real-time environmental condition monitoring data, the model parameters are dynamically adjusted through the strategy optimization module of the reinforcement learning algorithm to obtain the optimized wave force calculation model parameters;

[0024] S5: Based on the optimized wave force calculation model parameters, the real-time wave force value is obtained by updating the calculation model parameter configuration and re-executing the wave force calculation.

[0025] Specifically, the real-time wave force calculation method for breakwaters achieves dynamic and accurate wave force calculation through the coordinated execution of multiple steps. First, a wave monitoring sensor array and a structural response monitoring sensor network are deployed on the breakwater. These sensors simultaneously acquire wave physical characteristic data and structural dynamic response data. The wave monitoring sensor array typically includes wave height and wave direction sensors to acquire wave parameters such as wave height and wave incidence angle in real time; the structural response monitoring sensor network includes strain gauges and accelerometers to monitor the strain and vibration response of the breakwater structure. Synchronous acquisition ensures temporal consistency and spatial coverage of the data, providing comprehensive and real-time input for subsequent calculations. The technical advantage of this step is that it overcomes the problems of single data sources and delayed updates in traditional methods. By synchronously acquiring data from multiple sources, it improves the integrity and reliability of the data, laying a solid foundation for accurate wave force calculation. Synchronous acquisition avoids the accumulation of errors caused by data asynchrony, ensuring that wave parameters and structural response data accurately reflect the actual state of the breakwater, thus supporting the accurate operation of subsequent models.

[0026] Next, based on the collected wave parameters and the initial wave force calculation model pre-stored in the computing unit, the initial wave force value is obtained by calling the model's basic parameters and executing the wave force calculation formula. The initial wave force calculation model is a mathematical model based on physical principles, whose basic parameters include constants such as seawater density and gravitational acceleration, as well as coefficients calibrated using historical data. The computing unit calls these parameters, substitutes them with the wave parameters, executes the formula calculation, and outputs the initial wave force value. The technical advantage of this step is that it provides a rapid, preliminary wave force assessment, providing fundamental support for real-time response. The initial model, using preset parameters and formulas, can quickly generate calculation results. Although errors may exist, it provides a starting point for subsequent optimization, avoiding the delays caused by relying entirely on real-time data calculations.

[0027] Then, based on the structural response data and initial wave force values, an error signal reflecting the magnitude of the calculation error is obtained through structural mechanics model transformation and numerical difference analysis. The structural mechanics model converts the structural response data (such as strain and vibration) into equivalent force data, thereby obtaining the measured structural response wave force values. Numerical difference analysis calculates the difference between the initial wave force values ​​and the measured structural response wave force values, generating an original error data sequence. Further processing using a digital filtering algorithm eliminates noise interference, resulting in a smooth error signal. The technical advantage of this step is the accurate identification of calculation deviations, providing a crucial basis for model optimization. Through structural mechanics model transformation, measured data are converted into comparable force values, and then the error is quantified through difference analysis, ensuring that the error signal accurately reflects the gap between the model calculation and the actual state, providing accurate input for subsequent parameter adjustments.

[0028] Furthermore, based on error signals and real-time environmental monitoring data, the strategy optimization module of the reinforcement learning algorithm dynamically adjusts the model parameters to obtain optimized wave force calculation model parameters. Environmental monitoring data includes wind speed, wind direction, water temperature, and salinity, which are integrated using a multi-source data fusion processing algorithm to form a comprehensive environmental state input. The strategy optimization module of the reinforcement learning algorithm evaluates the parameter adjustment effect through a reward function based on the error signals and environmental state data, generates model parameter adjustment instructions, and then corrects the key parameters in the initial wave force calculation model. The technical effect of this step is that it achieves adaptive optimization of the model, significantly improving computational accuracy and dynamic adaptability. Through continuous learning and adjustment, the reinforcement learning algorithm enables the model to adapt to environmental changes, reducing error accumulation and ensuring high accuracy in wave force calculations even under complex conditions.

[0029] Finally, based on the optimized wave force calculation model parameters, the wave force calculation is re-executed by updating the model parameter configuration to obtain real-time wave force values. The calculation unit updates the model configuration according to the optimized parameters, re-runs the wave force calculation formula, outputs the corrected wave force values, and transmits them to the monitoring system through the data output interface. The technical effect of this step is to achieve real-time updating of model parameters and accurate output of wave forces, ensuring the timeliness and reliability of the calculation results. Parameter updates enable the model to reflect environmental changes and structural conditions in real time, and recalculation avoids error solidification, thereby outputting wave force data that is closer to the true value and supporting timely safety decisions.

[0030] Overall, this method achieves dynamic precision and adaptive adjustment in wave force calculation through a closed-loop collaboration of data acquisition, model calculation, error analysis, parameter optimization, and real-time output. Its technical effects are reflected in several aspects: First, simultaneous acquisition of multi-source data ensures the comprehensiveness and consistency of input information; second, initial model calculation provides a foundation for rapid response; third, error analysis accurately identifies deviations and provides a basis for optimization; fourth, reinforcement learning drives dynamic parameter adjustment, improving model adaptability; and finally, real-time output ensures the timeliness and usability of calculation results. These effects work together to overcome the problems of inaccurate risk assessment and delayed response in dynamic environments using traditional methods, thereby enhancing the initiative and reliability of breakwater safety control.

[0031] In some implementations, S1 includes:

[0032] S1.1: Based on the wave height sensing device installed on the wave-facing side of the breakwater, the raw wave height monitoring data is obtained by collecting wave surface elevation change data in real time.

[0033] S1.2: Based on the wave direction detection device installed on the side of the breakwater, the original wave direction monitoring data is obtained by measuring the wave propagation direction angle in real time;

[0034] S1.3: Based on a distributed strain sensing network embedded inside the concrete structure of the breakwater, structural strain response data is obtained by continuously monitoring the structural deformation strain distribution;

[0035] S1.4: Based on the original wave height monitoring data, original wave direction monitoring data, and structural strain response data, wave parameters and structural response data are obtained through data quality verification and format standardization.

[0036] This embodiment further refines the process of acquiring wave parameters and structural response data, ensuring the accuracy and completeness of the data through specific steps. First, based on wave height sensors deployed on the wave-facing side of the breakwater, raw wave height monitoring data is obtained by real-time acquisition of wave surface elevation changes. Wave height sensors typically employ radar or ultrasonic sensors to continuously monitor changes in water surface position, thereby acquiring wave height information. The technical advantage of this step is that it provides direct and continuous wave height measurement, avoiding errors caused by indirect estimation. Real-time acquisition ensures the timeliness of wave height data, while direct measurement reduces intermediate conversion steps, improving the authenticity and reliability of the data and providing high-quality input for subsequent wave parameter calculations.

[0037] Secondly, based on wave direction detection devices installed on the side of the breakwater, raw wave direction monitoring data is obtained by measuring the wave propagation direction angle in real time. The wave direction detection device may include a directional wave height sensor or an image processing system, which determines the direction angle by analyzing the wave propagation characteristics. The technical advantage of this step is that it accurately captures the wave propagation direction, which is crucial for the incident angle parameter in wave force calculation. Direction data directly affects the incident angle variable in the wave force calculation formula; real-time measurement ensures the accuracy of this parameter, thereby reducing calculation deviations caused by direction errors and improving the overall model accuracy.

[0038] Then, based on a distributed strain sensing network embedded within the breakwater's concrete structure, structural strain response data is obtained by continuously monitoring the structural deformation and strain distribution. The distributed strain sensing network typically employs fiber optic grating sensors or resistance strain gauges, arranged along key sections of the breakwater to capture strain changes under different loads. The technical advantage of this step is that it enables real-time monitoring of the structure's internal state, allowing for the timely detection of potential damage or abnormal deformation. Strain data directly reflects the structural stress state; continuous monitoring captures the response under dynamic loads, providing reliable input for structural mechanics model conversion and thus supporting the accurate generation of error signals.

[0039] Finally, based on the original wave height monitoring data, original wave direction monitoring data, and structural strain response data, wave parameters and structural response data were obtained through data quality verification and format standardization. Data quality verification included outlier detection, consistency checks, and missing value handling to ensure the data was free of significant errors; format standardization converted data from different sources into a unified format, facilitating subsequent calculations. The technical effect of this step is to improve data consistency and usability, and reduce model errors caused by data quality issues. Verification and standardization eliminated noise and inconsistencies during data acquisition, enabling wave parameters and structural response data to be seamlessly integrated into the calculation process, supporting the accurate execution of subsequent steps.

[0040] Overall, by refining the data acquisition and processing steps, the accuracy, real-time performance, and consistency of wave parameters and structural response data were ensured. The technical benefits are multifaceted: direct measurement of wave height and direction data reduces indirect errors; continuous monitoring of strain data provides a true reflection of the structural state; and data verification and standardization improve the overall data quality. These effects collectively enhance the reliability of the initial data input, laying a solid foundation for subsequent wave force calculations, thereby indirectly improving the accuracy and adaptability of the entire method.

[0041] In some implementations, S2 includes:

[0042] S2.1: Based on wave parameters, obtain the model calculation parameters corresponding to the current working condition by accessing the model parameter database, and obtain the initial model calculation parameter set;

[0043] S2.2: Based on wave parameters and the initial model calculation parameter set, the uncorrected wave force value is obtained by performing mathematical operations in the wave force calculation model.

[0044] S2.3: Based on the uncorrected wave force values, the initial wave force values ​​are obtained through data rationality verification and output format standardization.

[0045] This embodiment further refines the calculation process of the initial wave force value, ensuring the accuracy and efficiency of the model calculation through specific steps. First, based on wave parameters, the system retrieves the model calculation parameters corresponding to the current working condition from the model parameter database, obtaining the initial set of model calculation parameters. The model parameter database stores parameter values ​​under different working conditions, such as seawater density, gravitational acceleration, and empirical coefficients, which are determined based on historical data or experimental calibration. When accessing the database, the system matches the most relevant working condition based on real-time wave parameters (such as wave height and wave incidence angle) and calls the corresponding parameters. The technical advantage of this step is that it provides contextualized parameter input, ensuring the relevance of the model calculation. By dynamically matching working conditions, the model parameters can adapt to changes in the actual environment, avoiding errors caused by fixed parameters, thereby improving the accuracy of the initial calculation.

[0046] Secondly, based on wave parameters and the initial model calculation parameter set, uncorrected wave force values ​​are obtained by performing mathematical operations within the wave force calculation model. These mathematical operations typically involve solving physical formulas, such as substituting wave height, wave incidence angle, and parameters into formulas for numerical calculations. The technical advantage of this step is the rapid generation of preliminary wave force assessments, providing a foundation for real-time response. Utilizing predefined models and parameters, the calculation process is efficient and direct. Although the results may require subsequent correction, it provides necessary intermediate outputs for the entire process, supporting real-time requirements.

[0047] Then, based on the uncorrected wave force values, initial wave force values ​​are obtained through data reasonableness checks and output format normalization. Data reasonableness checks include range checks, trend analysis, and consistency verification, ensuring, for example, that the wave force values ​​are within physically possible ranges and avoiding abnormal outputs. Output format normalization converts the calculation results into a standard data structure, facilitating subsequent processing. The technical effect of this step is to improve data reliability and compatibility. The checking process filters out obvious errors, and normalization ensures smooth data transmission within the system, thereby reducing the accumulation of errors in subsequent steps due to input problems.

[0048] Overall, by refining the steps of model parameter acquisition, mathematical calculations, and data post-processing, the accuracy and usability of the initial wave force calculations are ensured. The technical benefits are reflected in several aspects: dynamic parameter matching adapts the model to actual working conditions; mathematical calculations provide rapid preliminary results; and data verification and standardization improve output quality. These effects work together to make the initial calculation steps both efficient and reliable, providing a stable foundation for subsequent error analysis and optimization, thereby enhancing the overall performance and practicality of the method.

[0049] In some implementations, S3 includes:

[0050] S3.1: The structural response data is converted into force data through a structural mechanics model to obtain the measured structural response wave force values;

[0051] S3.2: Based on the initial wave force value and the measured structural response wave force value, the original error data sequence is obtained through point-by-point numerical comparison and difference calculation;

[0052] S3.3: Based on the original error data sequence, noise interference is eliminated through a digital filtering algorithm to obtain the error signal.

[0053] This embodiment further refines the error signal generation process, ensuring the accuracy and stability of error calculation through specific steps. First, the structural response data is converted into force data using a structural mechanics model, yielding measured wave force values ​​based on the structural response. The structural mechanics model, based on the principles of materials mechanics and structural dynamics, converts strain and vibration response data into equivalent wave force values, for example, through stress-strain relationships or vibration frequency analysis. The technical advantage of this step is that it transforms measured data into directly comparable force values, providing a benchmark for error analysis. The structural mechanics model utilizes physical principles to ensure the accuracy of the conversion, enabling the measured force values ​​to accurately reflect the actual stress state of the breakwater, thereby supporting reliable error calculation.

[0054] Secondly, based on the initial wave force values ​​and the measured structural response wave force values, a sequence of original error data is obtained through point-by-point numerical comparison and difference calculation. Point-by-point numerical comparison calculates the difference between the initial force value and the measured force value at each time point or spatial point, generating a series of error data. The technical effect of this step is to quantify the deviation between the model calculation and the actual state. Through point-to-point comparison, the error data can accurately capture the spatiotemporal distribution of the calculation error, avoiding local problems that may be masked by overall averaging, thus providing a detailed basis for subsequent optimization.

[0055] Then, based on the original error data sequence, noise interference is eliminated through digital filtering algorithms to obtain the error signal. Digital filtering algorithms may include low-pass filtering or moving average processing to smooth the data sequence and remove high-frequency noise or random fluctuations. The technical effect of this step is to improve the stability and usability of the error signal. Filtering preserves the trend component of the error and eliminates noise during acquisition or transmission, making the error signal more clearly reflect the true deviation, thereby ensuring the accuracy and effectiveness of subsequent parameter adjustments.

[0056] Overall, by refining the structural mechanics model transformation, numerical comparison, and filtering steps, the accurate generation of error signals was ensured. The technical benefits are multifaceted: the structural mechanics transformation provides a reliable measured benchmark; point-by-point comparison quantifies detailed deviations; and filtering improves signal quality. These effects collectively enable the error signal to accurately reflect computational errors, providing crucial input for model optimization and indirectly enhancing the accuracy and robustness of the entire wave force calculation method.

[0057] In some implementations, S4 includes:

[0058] S4.1: Based on the error signal, the preset reward function calculation module performs evaluation calculations to obtain a reward signal that reflects the effect of parameter adjustment;

[0059] S4.2: Based on environmental status monitoring data and reward signals, a reinforcement learning decision network is used to analyze and make decisions, generating instructions for adjusting model parameters;

[0060] S4.3: Based on the model parameter adjustment command, the key parameters in the initial wave force calculation model are corrected through the parameter update algorithm to obtain the updated model parameter set;

[0061] S4.4: Based on the updated model parameter set, the optimized wave force calculation model parameters are obtained through parameter validity verification.

[0062] This embodiment further refines the dynamic adjustment process of model parameters, ensuring the efficiency and accuracy of optimization through specific steps. First, based on the error signal, a preset reward function calculation module performs evaluation calculations to obtain a reward signal reflecting the effect of parameter adjustment. The reward function calculation module evaluates the effect of the current parameter adjustment based on the magnitude and trend of the error signal; for example, a positive reward is given if the error decreases, and a negative reward is given if the error increases. The technical effect of this step is to quantify the quality of parameter adjustment, providing feedback for reinforcement learning. The reward signal transforms error information into a learnable signal, enabling the reinforcement learning algorithm to optimize its strategy based on historical performance, thereby guiding the direction and magnitude of subsequent parameter adjustments.

[0063] Secondly, based on environmental condition monitoring data and reward signals, a reinforcement learning decision network is used to analyze and make decisions, generating instructions for adjusting model parameters. Reinforcement learning decision networks typically employ deep networks or policy gradient methods, outputting parameter adjustment actions based on environmental condition data (such as wind speed and water temperature) and reward signals. The technical advantage of this step is that it achieves intelligent parameter optimization. Through learning and experience accumulation, the decision network can adapt to complex environmental changes and generate reasonable adjustment instructions, thereby reducing human intervention and improving the automation level and accuracy of optimization.

[0064] Then, based on the model parameter adjustment instructions, the key parameters in the initial wave force calculation model are corrected using a parameter update algorithm, resulting in an updated set of model parameters. The parameter update algorithm may include gradient descent or stochastic optimization methods, gradually modifying the model parameters according to the adjustment instructions. The technical advantage of this step is that it enables dynamic adaptation of the model. The update algorithm ensures smooth and effective parameter adjustment, avoiding system instability caused by abrupt changes, thus allowing the model to continuously adapt to new data and improve long-term performance.

[0065] Finally, based on the updated model parameter set, the optimized wave force calculation model parameters are obtained through parameter validity verification. Parameter validity verification includes range checking, consistency testing, and convergence evaluation to ensure that the parameters are within a reasonable range and compatible with the model. The technical effect of this step is to guarantee the credibility of the optimized parameters. The verification process filters out invalid or abnormal parameters, preventing erroneous adjustments from affecting the overall calculation, thereby ensuring the reliability and practicality of the optimization results.

[0066] Overall, by refining the reward calculation, decision generation, parameter update, and validation steps, the accuracy and efficiency of dynamic adjustment of model parameters are ensured. The technical effects are reflected in several aspects: the reward signal provides clear optimization guidance; the decision network achieves intelligent adjustment; the update algorithm ensures smooth correction; and the validation steps guarantee reliable results. These effects collectively enable the model to adapt to environmental changes, significantly improving the accuracy and robustness of wave force calculation, thereby enhancing the practical value of the entire method.

[0067] In some implementations, S4.2 includes:

[0068] S4.2.1: Based on environmental status monitoring data, an environmental status feature vector is obtained through feature extraction and vectorization encoding.

[0069] S4.2.2: Based on the environmental state feature vector and reward signal, a parameter adjustment action vector is generated through policy evaluation and value function calculation of a deep learning network;

[0070] S4.2.3: Based on the parameter adjustment action vector, the model parameter adjustment instructions are obtained through decoding the action-parameter mapping relationship.

[0071] This embodiment further refines the analysis and decision-making process of reinforcement learning decision networks, achieving intelligent parameter adjustment through three sub-steps: feature extraction, policy evaluation, and instruction decoding. Based on environmental state monitoring data, environmental state feature vectors are obtained through feature extraction and vectorization encoding. Environmental state monitoring data includes multi-dimensional information such as wind speed, wind direction, water temperature, and salinity. The feature extraction process identifies key features related to wave force calculation, such as the correlation between wind speed and wave energy, and the influence of water temperature on seawater density. Vectorization encoding transforms these features into numerical vector forms, preserving the spatial and temporal relationships in the original data. The technical effect of this processing is to transform complex environmental data into standard input that machines can understand, providing structured information for subsequent decision-making. Feature extraction filters out redundant information and highlights key factors, while vectorization encoding ensures data format uniformity, enabling reinforcement learning networks to efficiently process multi-dimensional inputs, thereby improving the accuracy and efficiency of decision-making.

[0072] Based on environmental state feature vectors and reward signals, parameter adjustment action vectors are generated through policy evaluation and value function calculation using a deep learning network. Deep learning networks typically employ an execution-evaluation collaborative architecture, where the policy network evaluates the optimal action in the current state, and the value function network predicts long-term returns. The reward signal serves as the training signal to guide network optimization, while the environmental state feature vector provides the decision-making basis. The network calculates the output action probability distribution through forward propagation, ultimately generating the parameter adjustment action vector. The technical advantage of this step lies in achieving intelligent decision-making based on deep learning, which comprehensively considers both environmental state and historical performance. The deep learning network captures the complex relationship between environmental features and parameter adjustment through nonlinear mapping, and the value function ensures that the decision considers not only immediate rewards but also long-term effects, thus making the generated adjustment actions more forward-looking and adaptive.

[0073] Based on parameter adjustment action vectors, the model parameter adjustment instructions are obtained through action-parameter mapping relationship decoding. The action-parameter mapping relationship is a pre-established correspondence rule that transforms abstract action vectors into specific model parameter modifications, such as adjusting coefficients or exponents in a wave force calculation model. The decoding process parses the meaning of the action vectors according to the mapping table or transformation function, generating executable instructions. The technical effect of this step is to transform machine learning output into engineering-usable control signals. Mapping relationship decoding builds a bridge between intelligent decision-making and concrete implementation, ensuring that the abstract results output by the neural network can be accurately translated into actual adjustments to the computational model, maintaining the consistency and operability of the entire optimization process.

[0074] These sub-steps together constitute a complete intelligent decision-making chain. Feature extraction ensures input quality, policy evaluation provides optimal decisions, and action decoding realizes instruction conversion. Through the construction of environmental state feature vectors, the system can comprehensively perceive changes in external conditions; through policy evaluation via deep learning networks, the system possesses autonomous learning and optimization capabilities; and through action-parameter mapping relationship decoding, the system ensures the executability of decision results. The entire process ensures that model parameter adjustment considers both real-time environmental characteristics and historical experience learning, while guaranteeing the accurate transmission of adjustment instructions. This significantly improves the environmental adaptability and prediction accuracy of the wave force calculation model, providing a more intelligent solution for breakwater safety monitoring.

[0075] In some implementations, S4.3 includes:

[0076] S4.3.1: Based on the model parameter adjustment instructions, the parameter smoothing adjustment amount is obtained by smoothing through a sliding window averaging algorithm;

[0077] S4.3.2: Based on the parameter smoothing adjustment, the coefficients in the initial wave force calculation model are gradually updated through the parameter iterative update formula to obtain a new set of model parameters;

[0078] S4.3.3: Based on the new set of model parameters, the updated set of model parameters is obtained through parameter boundary constraint checks and validity verification.

[0079] This embodiment refines the model parameter update process, ensuring the stability and effectiveness of parameter adjustment through three sub-steps: smoothing, iterative update, and boundary constraints. Based on the model parameter adjustment instructions, a sliding window averaging algorithm is used for smoothing to obtain the smoothed parameter adjustment amount. The sliding window averaging algorithm performs a weighted average of the parameter adjustment instructions over time, using a fixed-size window to cover the most recent multiple adjustment instructions and calculating their average as the final adjustment amount. The technical effect of this process is to eliminate random fluctuations and anomalous jumps during parameter adjustment. Through smoothing in the time dimension, the sliding window averaging filters out short-term noise interference, preserves the long-term trend of the adjustment instructions, and prevents drastic changes in parameters due to a single anomalous instruction, thereby maintaining the stability of model operation.

[0080] Based on the parameter smoothing adjustment, the coefficients in the initial wave force calculation model are progressively updated using a parameter iterative update formula to obtain a new set of model parameters. The parameter iterative update formula employs an incremental learning approach, incorporating the smoothing adjustment into the existing parameters with appropriate step sizes, such as through weighted averaging or gradient descent methods. The technical advantage of this step is that it achieves a smooth transition and continuous optimization of the parameters. Iterative updates avoid abrupt parameter changes, allowing the model parameters to gradually approach their optimal values ​​through multiple adjustments with small step sizes. This ensures learning efficiency while preventing overfitting or oscillations, enabling the model to smoothly adapt to environmental changes.

[0081] Based on the new set of model parameters, an updated set of model parameters is obtained through parameter boundary constraint checks and validity verification. Boundary constraint checks ensure that parameter values ​​are within physically reasonable ranges; for example, some coefficients should be positive or within specific intervals. Validity verification confirms the rationality of the new parameter set through simulation calculations or consistency checks. The technical effect of this step is to guarantee the physical meaning and computational stability of the optimized parameters. Boundary constraints prevent parameters from exceeding reasonable ranges, thus preventing model distortion, and validity verification ensures that the parameter combinations are mathematically and physically valid, thereby guaranteeing the reliability and practicality of the wave force calculation model.

[0082] These three sub-steps form a complete parameter update guarantee mechanism. Smoothing ensures input quality, iterative updates control the optimization process, and boundary constraints guarantee output quality. The sliding window averaging algorithm provides a stable input source for parameter updates, avoiding interference from abnormal commands; the iterative parameter update formula achieves incremental optimization, balancing learning speed and stability; parameter boundary constraint checks and validity verification serve as quality control steps, ensuring the reliability of the final parameter set. Through this meticulous design, the model parameter update process maintains both the flexibility of adaptive learning and the robustness required for engineering applications, enabling the wave force calculation model to maintain high accuracy and reliability during long-term operation, providing continuous and accurate data support for breakwater structural safety assessment.

[0083] In some implementations, S5 includes:

[0084] S5.1: Based on the optimized wave force calculation model parameters, the original model parameter settings are replaced by the model parameter configuration update operation to obtain the optimized wave force calculation model.

[0085] S5.2: Based on the wave parameters and the optimized wave force calculation model, re-execute the wave force calculation process to obtain the corrected wave force value;

[0086] S5.3: Based on the corrected wave force value, the data is transmitted to the monitoring system through the data output interface to obtain the real-time wave force value.

[0087] This embodiment details the final output stage of wave force calculation, ensuring accurate transmission and application of the calculation results through three sub-steps: model update, recalculation, and data transmission. Based on the optimized wave force calculation model parameters, the original model parameter settings are replaced through a model parameter configuration update operation, resulting in a parameter-optimized wave force calculation model. The model parameter configuration update operation loads the new parameter set into the runtime environment of the calculation system, overwriting the previous parameter settings. This process must ensure atomicity to avoid parameter inconsistencies during the update. The technical effect is the realization of real-time optimization and version management of the calculation model. Accurate parameter replacement ensures that the calculation model always uses the latest optimized parameters, avoiding calculation errors caused by mixing old and new parameters, and enabling wave force calculations to promptly reflect environmental changes and model optimization results.

[0088] Based on the wave parameters and the optimized wave force calculation model, the wave force calculation process is re-executed to obtain corrected wave force values. Re-execution refers to running the wave force calculation process completely once using the updated model and the latest wave parameter data, including formula solving, numerical calculation, and result generation. The technical benefit is obtaining accurate wave force values ​​based on the latest model and data. Recalculation ensures that all optimization effects are fully applied, avoiding the accumulation of errors that may result from using outdated intermediate results. This ensures that the output wave force values ​​reflect both real-time wave conditions and incorporate the model optimization results, significantly improving the accuracy of the calculation.

[0089] Based on the corrected wave force values, the data is transmitted to the monitoring system via a data output interface to obtain real-time wave force values. The data output interface is responsible for converting the calculation results into a standard format recognizable by the monitoring system and transmitting the data to the monitoring center via a communication protocol. The technical advantage lies in achieving seamless integration between the calculation system and the monitoring system. Standardized data output ensures that the monitoring system can accurately receive and analyze wave force data, providing timely and reliable information input for subsequent safety warnings and decision support, forming a complete closed loop from calculation to application.

[0090] These steps collectively constitute the output guarantee system for wave force calculations. Model updates ensure the up-to-dateness of the computational foundation, recalculation guarantees the accuracy of the results, and data transmission realizes application value. Updating model parameter configurations enables the practical application of optimization results; re-executing the wave force calculation process ensures that the calculation results are based on the latest model and data; and data output interface transmission establishes an information channel between the calculation system and the monitoring system. Through this complete process, wave force calculations not only achieve their own optimization and upgrade, but more importantly, transform the optimization results into practically usable monitoring data. This provides real-time and accurate data support for breakwater safety status assessment and risk early warning, ultimately improving the efficiency and reliability of the entire port safety control system.

[0091] In some implementation methods, the acquisition of environmental status monitoring data includes:

[0092] S4a: Based on the wind speed and direction monitoring device installed on the upper structure of the breakwater, environmental wind field data including wind speed and direction is obtained by collecting wind field characteristic data in real time;

[0093] S4b: Based on the temperature and salinity monitoring probes deployed in the waters in front of the breakwater, water temperature and salinity monitoring data are obtained by continuously monitoring the physical properties of the water.

[0094] S4c: Based on environmental wind field data, water temperature and salinity monitoring data, environmental status monitoring data is obtained through a multi-source data fusion processing algorithm.

[0095] This embodiment details the process of acquiring environmental condition monitoring data, ensuring the comprehensiveness and accuracy of environmental information through three sub-steps: wind field monitoring, water body monitoring, and data fusion. Based on a wind speed and direction monitoring device installed on the upper structure of the breakwater, environmental wind field data, including wind speed and direction, is obtained by real-time acquisition of wind field characteristic data. The wind speed and direction monitoring device typically employs an ultrasonic anemometer or a mechanical wind sensor to continuously measure wind speed and direction parameters. The technical advantage lies in acquiring key meteorological elements that directly affect wave generation and propagation. Wind field data is a crucial input parameter for wave force calculation; real-time acquisition ensures the capture of dynamic changes in the wind field, accurately reflecting the effect of wind on waves and providing reliable meteorological input for wave force calculation.

[0096] Based on temperature and salinity monitoring probes deployed in the waters in front of the breakwater, water temperature and salinity data are obtained by continuously monitoring the physical properties of the water. The temperature and salinity monitoring probes employ CTD sensors and are deployed at representative locations in the waters in front of the breakwater to continuously measure the water's temperature and salinity parameters. The technical advantage lies in obtaining key environmental parameters that affect the physical properties of seawater. Water temperature affects seawater density and viscosity, while salinity affects seawater density and buoyancy. These parameters indirectly affect wave propagation characteristics and their interaction with structures. Continuous monitoring ensures that changing trends in the aquatic environment can be captured, providing a more comprehensive environmental context for wave force calculations.

[0097] Based on environmental wind field data and water temperature and salinity monitoring data, environmental status monitoring data is obtained through a multi-source data fusion processing algorithm. This algorithm performs time synchronization, spatial registration, and weighted fusion of data from different sensors to generate a unified environmental status monitoring dataset. The technical effect is to form a comprehensive and consistent description of the environmental status. Data fusion eliminates the temporal and spatial inconsistencies of data from different sensors, and weighted fusion highlights the contribution of each environmental element. The generated environmental status monitoring data can more completely reflect the comprehensive environmental conditions of the breakwater, providing high-quality environmental input for subsequent wave force calculations and model optimization.

[0098] These environmental monitoring components collectively construct a multi-dimensional data acquisition network. Wind field monitoring captures the impact of the atmospheric environment, water body monitoring acquires marine environmental parameters, and data fusion integrates and enhances the information. Wind speed and direction monitoring devices provide direct measurement of wave generation driving forces; temperature and salinity monitoring probes acquire key parameters of water body physical properties; and multi-source data fusion processing algorithms integrate scattered environmental information into unified monitoring data. Through this environmental monitoring system, the system can comprehensively perceive the complex environmental conditions of the breakwater, accurately capture changes in key parameters such as wind, water temperature, and salinity, and provide rich and reliable environmental background data for wave force calculation. This enables the calculation model to better understand and respond to actual environmental conditions, ultimately improving the environmental adaptability and prediction accuracy of the entire real-time wave force calculation system.

[0099] Example 2

[0100] like Figure 2 As shown, in a second aspect, the present invention proposes a real-time wave force calculation system for breakwaters. The system employs the method provided in any of the above embodiments, and the system includes:

[0101] The sensor array module is used to obtain wave parameters including wave height and wave incidence angle, and structural response data including structural strain and vibration response, by synchronously collecting wave physical characteristic data and structural dynamic response data based on the wave monitoring sensor array and structural response monitoring sensor network installed on the breakwater.

[0102] The initial calculation module is used to obtain the initial wave force value by calling the basic parameters of the model and executing the wave force calculation formula based on the initial wave force calculation model pre-stored in the calculation unit.

[0103] The error analysis module is used to obtain an error signal reflecting the magnitude of the calculation error based on structural response data and initial wave force values, through structural mechanics model transformation and numerical difference analysis.

[0104] The reinforcement learning optimization module is used to dynamically adjust the model parameters based on error signals and real-time environmental condition monitoring data, through the strategy optimization module of the reinforcement learning algorithm, to obtain the optimized wave force calculation model parameters.

[0105] The real-time calculation module is used to obtain real-time wave force values ​​by updating the calculation model parameter configuration and re-executing the wave force calculation based on the optimized wave force calculation model parameters.

[0106] This system corresponds to the method provided in Embodiment 1 above, and will not be described in detail here.

[0107] 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 real-time calculation of wave force on a breakwater, characterized in that, include: S1: Based on the wave monitoring sensor array and structural response monitoring sensor network installed on the breakwater, wave parameters including wave height and wave incident angle and structural response data including structural strain and vibration response are obtained by synchronously collecting wave physical characteristic data and structural dynamic response data. S2: Based on wave parameters and the initial wave force calculation model pre-stored in the calculation unit, the initial wave force value is obtained by calling the basic parameters of the initial wave force calculation model and executing the mathematical operation process in the initial wave force calculation model; S3: Based on structural response data and initial wave force values, an error signal reflecting the magnitude of the calculation error is obtained through structural mechanics model transformation and numerical difference analysis; wherein, S3 includes: S3.1: The structural response data is converted into force data through a structural mechanics model to obtain the measured structural response wave force values; S3.2: Based on the initial wave force value and the measured structural response wave force value, the original error data sequence is obtained by comparing and calculating the differences between the initial wave force value and the measured structural response wave force value at each time point or spatial point. S3.3: Based on the original error data sequence, noise interference is eliminated through a digital filtering algorithm to obtain the error signal; S4: Based on error signals and real-time environmental condition monitoring data, the parameters of the initial wave force calculation model are dynamically adjusted through the strategy optimization module of the reinforcement learning algorithm to obtain the optimized wave force calculation model parameters; wherein, S4 includes: S4.1: Based on the error signal, the preset reward function calculation module performs evaluation calculations to obtain a reward signal that reflects the effect of parameter adjustment; S4.2: Based on environmental status monitoring data and reward signals, a reinforcement learning decision network is used to analyze and make decisions, generating instructions for adjusting model parameters; S4.3: Based on the model parameter adjustment command, the key parameters in the initial wave force calculation model are corrected through the parameter update algorithm to obtain the updated model parameter set; S4.4: Based on the updated model parameter set, the optimized wave force calculation model parameters are obtained through parameter validity verification; S4.2 includes: S4.2.1: Based on environmental status monitoring data, an environmental status feature vector is obtained through feature extraction and vectorization encoding. S4.2.2: Based on the environmental state feature vector and reward signal, a parameter adjustment action vector is generated through policy evaluation and value function calculation of a deep learning network; S4.2.3: Based on the parameter adjustment action vector, the model parameter adjustment instructions are obtained through decoding the action-parameter mapping relationship; S5: Based on the optimized wave force calculation model parameters, the real-time wave force value is obtained by updating the parameter configuration of the initial wave force calculation model and re-executing the wave force calculation.

2. The method for real-time calculation of wave force on a breakwater according to claim 1, characterized in that, S1 includes: S1.1: Based on the wave height sensing device installed on the wave-facing side of the breakwater, the raw wave height monitoring data is obtained by collecting wave surface elevation change data in real time. S1.2: Based on the wave direction detection device installed on the side of the breakwater, the original wave direction monitoring data is obtained by measuring the wave propagation direction angle in real time; S1.3: Based on a distributed strain sensing network embedded inside the concrete structure of the breakwater, structural strain response data is obtained by continuously monitoring the structural deformation strain distribution; S1.4: Based on the original wave height monitoring data, original wave direction monitoring data, and structural strain response data, wave parameters and structural response data are obtained through data quality verification and format standardization.

3. The method for real-time calculation of wave force on a breakwater according to claim 1, characterized in that, S2 include: S2.1: Based on wave parameters, obtain the model calculation parameters corresponding to the current working condition by accessing the model parameter database, and obtain the initial model calculation parameter set; S2.2: Based on wave parameters and the initial model calculation parameter set, the uncorrected wave force value is obtained by performing mathematical operations in the wave force calculation model. S2.3: Based on the uncorrected wave force values, the initial wave force values ​​are obtained through data rationality verification and output format standardization.

4. The method for real-time calculation of wave force on a breakwater according to claim 1, characterized in that, S4.3 includes: S4.3.1: Based on the model parameter adjustment instructions, the parameter smoothing adjustment amount is obtained by smoothing through a sliding window averaging algorithm; S4.3.2: Based on the parameter smoothing adjustment, the coefficients in the initial wave force calculation model are gradually updated through the parameter iterative update formula to obtain a new set of model parameters; S4.3.3: Based on the new set of model parameters, the updated set of model parameters is obtained through parameter boundary constraint checks and validity verification.

5. The method for real-time calculation of wave force on a breakwater according to claim 1, characterized in that, S5 include: S5.1: Based on the optimized wave force calculation model parameters, the original model parameter settings are replaced by the model parameter configuration update operation to obtain the optimized wave force calculation model. S5.2: Based on the wave parameters and the optimized wave force calculation model, re-execute the wave force calculation process to obtain the corrected wave force value; S5.3: Based on the corrected wave force value, the data is transmitted to the monitoring system through the data output interface to obtain the real-time wave force value.

6. The method for real-time calculation of wave force on a breakwater according to claim 1, characterized in that, The acquisition of environmental status monitoring data includes: S4a: Based on the wind speed and direction monitoring device installed on the upper structure of the breakwater, environmental wind field data including wind speed and direction is obtained by collecting wind field characteristic data in real time; S4b: Based on the temperature and salinity monitoring probes deployed in the waters in front of the breakwater, water temperature and salinity monitoring data are obtained by continuously monitoring the physical properties of the water. S4c: Based on environmental wind field data, water temperature and salinity monitoring data, environmental status monitoring data is obtained through a multi-source data fusion processing algorithm.

7. A real-time wave force calculation system for breakwaters, characterized in that, The system employs the method described in any one of claims 1 to 6, the system comprising: The sensor array module is used to obtain wave parameters including wave height and wave incidence angle, and structural response data including structural strain and vibration response, by synchronously collecting wave physical characteristic data and structural dynamic response data based on the wave monitoring sensor array and structural response monitoring sensor network installed on the breakwater. The initial calculation module is used to obtain the initial wave force value by calling the basic parameters of the initial wave force calculation model and executing the mathematical operation process in the initial wave force calculation model based on wave parameters and pre-stored in the calculation unit. The error analysis module is used to obtain an error signal reflecting the magnitude of the calculation error based on structural response data and initial wave force values, through structural mechanics model transformation and numerical difference analysis. The reinforcement learning optimization module is used to dynamically adjust the parameters of the initial wave force calculation model based on error signals and real-time collected environmental condition monitoring data, and obtain the optimized wave force calculation model parameters through the strategy optimization module of the reinforcement learning algorithm. The real-time calculation module is used to obtain real-time wave force values ​​by updating the parameter configuration of the initial wave force calculation model and re-executing the wave force calculation based on the optimized wave force calculation model parameters.

Images