A method and system for intelligent prediction of the stability of double-walled steel cofferdam structures

Abstract

This invention relates to the field of cofferdam stability prediction technology, and specifically to an intelligent prediction method and system for the stability of double-walled steel cofferdam structures. The method includes: acquiring the installation locations of sensors; arranging a sensor cluster according to the sensor locations; collecting data from the sensor cluster to obtain sampling information, including strain data, water flow velocity data, and soil pore water pressure data of the double-walled steel cofferdam structure; preprocessing the sampling information to obtain preprocessed sampling information; correcting a preset numerical model of the double-walled steel cofferdam based on the preprocessed sampling information to obtain a corrected numerical model; and predicting the stability level of the double-walled steel cofferdam structure based on the corrected numerical model to obtain the prediction result. This invention can detect potential safety hazards in advance, assisting engineers in taking timely measures to ensure the safety of construction and use of double-walled steel cofferdams and reduce accident risks.

Description

Technical Field

[0001] This invention relates to the field of cofferdam stability prediction technology, and more specifically, to an intelligent prediction method and system for the stability of double-walled steel cofferdam structures. Background Technology

[0002] Double-walled steel cofferdams serve as crucial temporary support structures for deep-water bridges, ports, and other engineering projects, and their stability directly determines construction safety and efficiency. Due to the complex and variable underwater environment (such as water flow impact, wave loads, and foundation softening), cofferdams may experience catastrophic accidents due to overturning, sliding, or seepage instability. Therefore, accurate stability prediction is a core requirement for engineering safety. Existing technologies primarily predict stability through theoretical mechanical analysis (static equilibrium equations), finite element numerical simulation (fluid-structure interaction simulation), and physical model tests (fluidized tank scale tests). However, these methods still fall short of accurately predicting the stability of underwater double-walled steel cofferdam structures. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent prediction method and system for the stability of double-walled steel cofferdam structures, so as to improve the above-mentioned problems.

[0004] To achieve the above objectives, the embodiments of this application provide the following technical solutions:

[0005] On the one hand, embodiments of this application provide an intelligent prediction method for the stability of a double-walled steel cofferdam structure, the method comprising:

[0006] Obtain the sensor's installation location;

[0007] Arrange the sensor cluster according to the installation location of the sensors;

[0008] Based on the data collected by the sensor cluster, sampling information is obtained, which includes strain data of the double-walled steel cofferdam structure, water flow velocity data, and soil pore water pressure data.

[0009] The sampled information is preprocessed to obtain preprocessed sampled information;

[0010] The pre-processed sampling information is used to correct the preset numerical model of the double-walled steel cofferdam, resulting in the corrected numerical model.

[0011] The stability level of the double-walled steel cofferdam structure was predicted based on the modified numerical model, and the prediction results were obtained.

[0012] Secondly, embodiments of this application provide an intelligent prediction system for the stability of a double-walled steel cofferdam structure, the system comprising:

[0013] The acquisition module is used to acquire the installation location of the sensor;

[0014] The first processing module is used to arrange the sensor cluster according to the installation location of the sensor;

[0015] The second processing module is used to obtain sampling information based on the data collected by the sensor cluster. The sampling information includes strain data of the double-walled steel cofferdam structure, water flow velocity data, and soil pore water pressure data.

[0016] The third processing module is used to preprocess the sampling information to obtain preprocessed sampling information;

[0017] The fourth processing module is used to correct the preset numerical model of the double-walled steel cofferdam based on the preprocessed sampling information to obtain the corrected numerical model.

[0018] The prediction module is used to predict the stability level of the double-walled steel cofferdam structure based on the modified numerical model, and obtain the prediction results.

[0019] Thirdly, embodiments of this application provide an intelligent prediction device for the stability of a double-walled steel cofferdam structure. The device includes a memory and a processor. The memory stores a computer program; the processor executes the computer program to implement the steps of the aforementioned intelligent prediction method for the stability of a double-walled steel cofferdam structure.

[0020] Fourthly, embodiments of this application provide a readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described intelligent prediction method for the stability of a double-walled steel cofferdam structure.

[0021] The beneficial effects of this invention are as follows:

[0022] This invention determines the sensor installation location before installing the sensor to collect data, thus obtaining sampling information. This solves the problem of low data quality due to the complex environment in which the double-walled steel cofferdam structure is located. The data is then preprocessed, and the preprocessed sampling information is used to correct the preset numerical model of the double-walled steel cofferdam, establishing a multi-physics digital twin model. Based on the multi-physics digital twin model, the stability level of the double-walled steel cofferdam structure can be predicted in complex underwater environments. This effectively improves the stability prediction accuracy of the double-walled steel cofferdam structure in complex environments, identifies potential safety hazards in advance, helps engineers take timely measures, ensures the safety of construction and use of the double-walled steel cofferdam, and reduces the risk of accidents.

[0023] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram of the intelligent prediction method for the stability of a double-walled steel cofferdam structure as described in an embodiment of the present invention.

[0026] Figure 2 This is a schematic diagram of the intelligent prediction system for the stability of a double-walled steel cofferdam structure as described in an embodiment of the present invention.

[0027] Figure 3 This is a schematic diagram of the intelligent prediction device for the stability of a double-walled steel cofferdam structure as described in an embodiment of the present invention.

[0028] The diagram is labeled as follows: 800, Intelligent prediction device for the stability of double-walled steel cofferdam structure; 801, Processor; 802, Memory; 803, Multimedia component; 804, I / O interface; 805, Communication component; 901, Acquisition module; 902, First processing module; 903, Second processing module; 904, Third processing module; 905, Fourth processing module; 906, Prediction module. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0030] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0031] Example 1:

[0032] This embodiment provides an intelligent prediction method for the stability of a double-walled steel cofferdam structure. It can be understood that a scenario can be set up in this embodiment, such as a scenario for predicting the stability of an underwater double-walled steel cofferdam structure.

[0033] See Figure 1 The figure shows that the method includes steps S1, S2, S3, S4, S5 and S6.

[0034] Step S1: Obtain the sensor's installation location;

[0035] Step S1 further includes steps S11, S12, and S13, which specifically include:

[0036] Step S11: Obtain the dynamic load model and structural information of the double-walled steel cofferdam;

[0037] In this step, the specific process of obtaining the dynamic load model is as follows: Data such as water flow velocity, wave height, and tidal level around the double-walled steel cofferdam are collected from various sensors. This data includes both historical and real-time data. The collected data is then processed to ensure its accuracy and completeness for subsequent analysis. Preprocessing of the collected data includes, but is not limited to, cleaning to remove outliers and noise. An LSTM model is built and trained using the preprocessed data to obtain the dynamic load prediction model.

[0038] Step S12: Determine the load of water flow on different parts of the double-walled steel cofferdam based on the dynamic load model;

[0039] In this step, the dynamic load model learns from a large amount of data to capture the changing patterns of dynamic loads and predict parameters such as the magnitude, direction, and frequency of water flow force, wave force, and tidal force in the future. This allows the model to determine the impact of dynamic loads such as water flow force, wave force, and tidal force on different parts of the double-walled steel cofferdam during actual operation.

[0040] Step S13: Determine the installation position of the sensor based on the load of the water flow on different parts of the double-walled steel cofferdam and the structural information of the double-walled steel cofferdam.

[0041] In this step, after clarifying the loads of water flow on different parts of the double-walled steel cofferdam, the characteristics of different loads and the resulting structural responses can be determined. For example, if the water flow force is predicted to have high-frequency periodic changes in a certain direction, it can be known that the double-walled steel cofferdam structure in that direction will bear large alternating stress, which is a critical stress area. Based on this, the key areas for sensor placement can be accurately located, avoiding blind placement and making sensor monitoring more targeted, ensuring the acquisition of the most critical structural state information. At the same time, based on the structural information of the double-walled steel cofferdam, its stress concentration areas can also be determined, such as the connection between the inner and outer steel plates and the horizontal truss and bulkheads, where stress concentration is prone to occur under dynamic loads due to the difference in stiffness of different components; and the area near the bottom of the riverbed, which not only bears large soil pressure but is also affected by water erosion and tidal forces. Based on this, the key areas for sensor placement can also be determined.

[0042] In this embodiment, the sensor placement is optimized based on the dynamic load model and the structural information of the double-walled steel cofferdam. This can effectively improve the quality of data acquisition in complex underwater environments and further improve the accuracy of stability prediction.

[0043] Step S2: Arrange the sensor cluster according to the installation location of the sensor;

[0044] In this step, the sensor cluster includes, but is not limited to, fiber optic strain gauges and MEMS tiltmeters for structural monitoring; acoustic Doppler current meters and miniature pressure sensors for flow field monitoring; and piezometers and distributed fiber optic thermometers for soil monitoring.

[0045] Step S3: Based on the data collected by the sensor cluster, obtain sampling information, which includes strain data of the double-walled steel cofferdam structure, water flow velocity data, and soil pore water pressure data.

[0046] Step S4: Preprocess the sampling information to obtain preprocessed sampling information;

[0047] Step S4 further includes steps S41, S42, and S43, which specifically include:

[0048] Step S41: Use Kalman filtering to process the water flow velocity data included in the sampled information to obtain the processed first information;

[0049] In this step, adaptive Kalman filtering is used to eliminate high-frequency vibration interference. Specifically: (1) Construct a state space model of the vibration system of the double-walled steel cofferdam, define state variables including key physical quantities such as structural displacement, velocity, and acceleration, and establish state equations and observation equations. Based on the original vibration signal from the sensor, output the initial state vector and covariance matrix; (2) Enter the prediction-update loop. In the prediction stage, predict the current state and covariance based on the state at the previous moment; in the update stage, combine real-time vibration data to correct the prediction results, calculate the residuals and monitor their statistical characteristics. By iterating, gradually approximate the real state and output the filtered state estimate and residuals; (3) When the residual variance exceeds the preset threshold, use maximum a posteriori estimation to optimize the noise covariance matrix online, so that it can adapt to non-steady-state high-frequency interference and ensure the positive definiteness of the matrix. Finally, extract the displacement from the filtered state to complete the signal reconstruction, verify the high-frequency noise suppression effect through frequency domain analysis, confirm the optimality of the filter by residual whitening test, and output the denoised vibration signal, which is the first information after processing.

[0050] Step S42: Use wavelet threshold denoising method to process the soil pore water pressure data included in the sampling information to obtain the processed second information;

[0051] In this step, a specific implementation method is as follows: Based on the characteristics of low-frequency gradual variation and random noise in the soil seepage pressure signal, Symlets wavelet is selected to perform wavelet decomposition on the soil pore water pressure data included in the sampling information, separating low-frequency approximation coefficients and high-frequency detail coefficients; Stein unbiased risk estimation is used to determine the global optimal threshold, and a layered adaptive threshold is calculated. A soft thresholding function is used to process the high-frequency detail coefficients, retaining the low-frequency approximation coefficients and suppressing the noise-dominated high-frequency components, resulting in processed coefficients; the processed coefficients are then subjected to inverse wavelet transform to reconstruct the signal, and drift estimation is calculated and baseline is corrected using moving average to obtain a denoised and drift-free soil pore water pressure signal. It should be noted that the calculation of the layered adaptive threshold is specifically as follows:

[0052] ;

[0053] In the above formula, where, This represents the adaptive threshold corresponding to the detail coefficients of the i-th layer. denoted as the standard deviation of the noise in the i-th layer, and N represents the number of samples of soil pore water pressure data.

[0054] Step S43: Obtain preprocessed sampling information based on the processed first information and the processed second information.

[0055] In this embodiment, Kalman filtering focuses on suppressing high-frequency vibration noise, while wavelet denoising eliminates low-frequency drift and complementarily covers the entire frequency band, significantly improving data quality and laying a reliable foundation for subsequent stability analysis.

[0056] Step S5: Correct the preset numerical model of the double-walled steel cofferdam based on the preprocessed sampling information to obtain the corrected numerical model.

[0057] Step S5 further includes steps S51, S52, S53, S54, and S55, which specifically include:

[0058] Step S51: Construct a state vector based on the preprocessed sampling information;

[0059] In this step, the state vector is defined to include structural stress, water flow velocity, and soil pore water pressure. The state vector is used to comprehensively describe the state of the double-walled steel cofferdam.

[0060] Step S52: Calculate the observation covariance matrix, which is used to measure the correlation and variability between state vectors;

[0061] In this step, the observation covariance matrix is ​​calculated, which measures the correlation and variability among the observed data. By analyzing the relationships between sensor data, the degree of influence of each observation on the model prediction is determined, providing a basis for subsequent updates to the state vector weights.

[0062] Step S53: Update the weights of the state vector according to the observation covariance matrix;

[0063] In this step, the Kalman gain is first calculated as follows:

[0064] ;

[0065] In the above formula, Indicates Kalman gain, The covariance matrix represents the prediction error of the numerical model, reflecting the uncertainty of the numerical model prediction; H is the observation matrix, used to map the model state variables to the observation space; R is the observation covariance matrix calculated based on the preprocessed sampling information acquired in real time. After calculating the Kalman gain, the weights of the state vector are updated according to the update formula, which is as follows:

[0066] ;

[0067] In the above formula, y is the state vector predicted by the numerical model, H is the actual observation vector, K is the Kalman gain, and x is the updated state vector. In this way, the updated state vector is closer to the actual situation, reducing the model error, and thus making the model's prediction of state variables such as structural stress, water flow velocity, and soil pore pressure of the double-walled steel cofferdam more consistent with the actual measurement situation.

[0068] Step S54: Use the particle swarm optimization algorithm to determine the hidden parameters and obtain the corrected multiphysics state parameters;

[0069] First, the particle swarm optimization (PSO) algorithm is initialized to determine the parameters to be inverted, including the soil elastic modulus and the residual stress of the steel. The sum of squared errors between the numerical model's predicted structural stress, water flow velocity, and soil pore pressure and the actual measured values ​​is used as the fitness function metric. The PSO algorithm then iterates, updating the velocity and position of each particle based on its historical best position and the global best position of the population, until a preset maximum number of iterations is met. This yields the values ​​of implicit parameters such as the soil elastic modulus and the residual stress of the steel. In this step, the implicit parameter values ​​obtained from the PSO algorithm are substituted into the model, combined with the state vector updated by the previous ensemble Kalman filtering, to dynamically adjust the model's parameters, thus obtaining the corrected multiphysics state parameters, including new structural stress, water flow velocity, and soil pore pressure. These corrected parameters more accurately reflect the actual state of the double-walled steel cofferdam, providing reliable data support for subsequent analysis and decision-making, and also providing accurate information for subsequent mesh adjustments, making the entire model closer to reality.

[0070] Step S54 further includes steps S541, S542, and S543, which specifically include:

[0071] Step S541: Obtain the maximum and minimum values ​​of the inertia weight;

[0072] Step S542: Optimize the inertia weight based on the maximum and minimum values ​​of the inertia weight to obtain the optimized inertia weight;

[0073] Step S543: Iteratively update the particle position based on the optimized inertia weight to obtain the optimal particle position and determine the hidden parameters.

[0074] In this embodiment, the optimized inertia weight is specifically as follows:

[0075] ;

[0076] In the above formula, and These represent the minimum and maximum values ​​of the inertia weight, respectively, used to limit the range of values ​​for the inertia weight and provide boundary conditions for the algorithm's search. This represents the fitness value of the i-th particle in the current iteration. This represents the average fitness value of particles in the current population. This represents the optimal fitness value of a particle in the current population, and the inertia weight is dynamically adjusted based on the particle's fitness value. In the stability prediction of double-walled steel cofferdams, when the difference between the particle's fitness value and the optimal fitness value is large, the inertia weight is larger. This allows the particle to jump out of the current local region and search in a larger space, avoiding getting trapped in local optima, thus making it more likely to find the globally optimal combination of model parameters and improving the accuracy of the stability prediction of double-walled steel cofferdams. When considering the influence of multiple complex factors on the stability of double-walled steel cofferdams, more reasonable parameters can be searched. At the same time, when the particle is close to the optimal solution region and the difference between the fitness value and the optimal fitness value becomes smaller, in the double-walled steel cofferdam model parameter correction, the particle can conduct a fine-tuned search locally and fine-tune the parameters to better fit the actual situation, improving the accuracy of the model's stability prediction of double-walled steel cofferdams and accurately reflecting their stability state.

[0077] Step S55: Adjust the grid resolution of the preset double-walled steel cofferdam numerical model according to the corrected multiphysics state parameters to obtain the corrected numerical model.

[0078] In this step, based on the corrected multiphysics state parameters, high-gradient regions requiring mesh refinement are identified in both the structural and soil fields. In the structural field, the focus is on locating stress concentration areas, such as welds and support nodes. These areas often exhibit stress concentration due to abrupt changes in structural shape or complex stress distribution. The intensity of stress change is quantified by calculating the stress gradient; a larger stress gradient indicates faster stress change in the region, requiring a finer mesh to accurately capture the stress distribution. For the soil field, a shear strain rate exceeding the threshold indicates more severe soil deformation in that region, necessitating a finer mesh to describe its mechanical behavior. After identifying high-gradient regions, an h-adaptive method is used for local mesh refinement, reducing the mesh size to 1 / 4 of its original size. This increases mesh resolution, thereby improving the computational accuracy in these regions and more accurately simulating the mechanical responses of the structure and soil in these areas.

[0079] After adjusting the grid resolution, a sample set containing over 1000 working conditions was generated, comprehensively covering different load combinations and boundary conditions to simulate various situations that a double-walled steel cofferdam might encounter. Then, a convolutional autoencoder was used to process the sample set, mapping high-dimensional data to a low-dimensional space, extracting low-dimensional features with a potential spatial dimension not exceeding 100, and constructing a reduced-order model. This reduced-order model was used to predict the stress, displacement, and seepage fields online. The online predictions of the stress, displacement, and seepage fields by the reduced-order model were compared with real-time sensor data to calculate the error. This method verifies the accuracy of the model's predictions in real time and determines its reliability. Once the model is deemed reliable, a corrected numerical model is obtained. It should be noted that, in terms of fluid-structure interaction, the submerged boundary method was used to handle the interaction between water flow and the structure, accurately simulating the force exerted by water flow on the double-walled steel cofferdam structure and the structure's response under water flow. In terms of solid-soil interaction, a contact algorithm was used to simulate the force transmission at the cofferdam-foundation interface, considering the supporting and constraining effects of the soil on the cofferdam.

[0080] Step S6: Predict the stability level of the double-walled steel cofferdam structure based on the modified numerical model, and obtain the prediction results.

[0081] Step S6 further includes steps S61, S62, S63, S64, and S65, which specifically include:

[0082] Step S61: Obtain physical field response data based on the modified numerical model;

[0083] Step S62: Calculate the deterministic safety factor based on the physical field response data;

[0084] In this step, the deterministic safety factor includes the overturning safety factor, the sliding safety factor, and the seepage stability factor. The specific calculation process for the overturning safety factor is as follows:

[0085] ;

[0086] In the above formula, W represents the self-weight of the double-walled steel cofferdam structure. This represents the horizontal distance from the line of action of the structure's self-weight to the overturning point. Indicates the frictional force of the base. This represents the horizontal distance from the line of action of the base friction force to the overturning point. and These represent water pressure and earth pressure, respectively, and H represents the vertical distance from the point of application of the water pressure to the bottom of the double-walled steel cofferdam structure.

[0087] The specific calculation process for the anti-slip safety factor is as follows:

[0088] ;

[0089] In the above formula, denoted by , c represents the soil cohesion, N represents the vertical resultant force, and T represents the horizontal load.

[0090] The specific calculation process for the seepage stability coefficient is as follows:

[0091] ;

[0092] In the above formula, Indicates the seepage stability coefficient. and These represent the actual hydraulic gradient and the critical hydraulic gradient, respectively.

[0093] Step S63: Generate a perturbed sample set using the Latin hypercube sampling method;

[0094] In practical engineering, uncertainties exist in environmental loads (such as wave height following a Weibull distribution and peak ground acceleration following the GR law), material parameters (soil elastic modulus normally distributed with a coefficient of variation of ±20%, weld strength Weibull distribution), and geometric errors (cofferdam installation misalignment uniformly distributed ±50mm). These uncertainties can affect the stability of engineering structures. To comprehensively evaluate the model's performance under uncertainties, 1000 parameter combinations were generated using Latin hypercube sampling. These combinations cover the parameter space with the lowest possible bias.

[0095] Step S64: Calculate the failure probability based on the set of disturbance samples;

[0096] Step S64 further includes steps S641, S642, S643, S644, and S645, which specifically include:

[0097] Step S641: Calculate the overturning safety factor and the slippage safety factor for each disturbance sample in the disturbance sample set;

[0098] Step S642: Obtain the first threshold information and the second threshold information;

[0099] Step S643: Determine whether the overturning safety factor corresponding to each disturbance sample is less than the first threshold information, and obtain the first judgment result;

[0100] Step S644: Determine whether the anti-slip safety factor corresponding to each disturbance sample is less than the second threshold information, and obtain the second judgment result;

[0101] Step S645: Calculate the failure probability based on the first judgment result and the second judgment result.

[0102] In this step, disturbance samples corresponding to an overturning safety factor less than the first threshold or a slip resistance safety factor less than the second threshold are taken as failure samples. The specific calculation process for the failure probability is as follows:

[0103] ;

[0104] In the above formula, Indicates the probability of failure. This represents the number of disturbance samples whose overturning safety factor is less than the first threshold or whose slippage safety factor is less than the second threshold. One disturbance sample corresponds to two safety factors. Both must be greater than the threshold information to be considered a failure sample. Any safety factor less than the threshold information is considered a failure sample in the calculation.

[0105] Step S65: Predict the stability level of the double-walled steel cofferdam structure based on the deterministic safety factor and the failure probability.

[0106] In this step, the prediction of the stability level of the double-walled steel cofferdam structure is specifically as follows:

[0107] ;

[0108] In the above formula, R is the overall risk value; , as well as These are weighting coefficients, determined through expert experience and historical accident data; and These are the actual calculated overturning safety factors and sliding safety factors; and These are the pre-set standard values ​​for the overturning safety factor and the pre-set standard values ​​for the slippage safety factor; , , , All are adjusted indices, among which, This is used to control the nonlinear relationship between the degree to which the overturning safety factor deviates from the standard value and the overall risk value. This is used to adjust the degree to which the anti-slip safety factor deviates from the standard value and its impact on the overall risk value. Used to reflect the sensitivity of an engineering project to probabilistic failure risks. Used to control the influence of the ratio of seepage gradient to critical seepage gradient on the overall risk value; Indicates the probability of failure; Indicates the seepage stability coefficient; , These represent the time decay coefficients corresponding to the overturning and slip resistance safety factors and the failure probability, respectively, determined based on the aging characteristics of engineering materials and environmental corrosion. In a specific implementation, when the comprehensive risk value is less than 0.3, it belongs to the normal level; when the comprehensive risk value is greater than or equal to 0.3 and less than 0.6, it belongs to the warning level; when the comprehensive risk value is greater than or equal to 0.6, it belongs to the emergency level. A hierarchical control strategy is generated based on the predicted stability level (normal / warning / emergency): when the predicted stability level is a warning, the sampling frequency of the sensor in the abnormal area is increased from 1Hz to 50Hz to focus on data acquisition; at the same time, the model is verified based on the re-acquired data, and then the risk location image and suggested measures are pushed to the engineer; when the predicted stability level is an emergency, the location and volume of the compartment that needs to be injected with water are calculated based on the buoyancy balance equation, and the calculation result is obtained; based on the calculation result, a control command is issued, which is used to adjust the opening of the solenoid valve to control the water injection volume.

[0109] It should be noted that the specific calculation process for the results is as follows:

[0110] ;

[0111] In the above formula, Indicates the calculation result. The density of water, This indicates the magnitude of the force corresponding to the additional moment required to balance the overturning moment. ,in, Overturning moment represents the difference between the overturning moment currently borne by the structure and the allowable overturning moment. Represents gravitational acceleration. This indicates the length of the lever arm.

[0112] This invention formulates differentiated control strategies for different risk levels. When it is in the early warning level (0.3≤R<0.6), by increasing the sensor sampling frequency, triggering model recalculation, and manual confirmation, it can more accurately capture abnormal data in the early stage of risk, deeply analyze risk trends, avoid misjudgment or omission, and prepare for risk response in advance. After entering the emergency level (R≥0.6), it immediately initiates counterweight decision-making to quickly and accurately intervene in high-risk conditions and maximize the safety of the engineering structure.

[0113] This invention integrates deterministic safety factors and failure probabilities to quantify risk values, fully utilizing these abundant data resources. This allows risk quantification to move beyond limitations of single data types or analytical methods, significantly improving its accuracy. Deterministic safety factors and probabilistic failure risks reflect structural risks from different perspectives. When the deterministic safety factor is close to but not yet reached a threshold, the probabilistic failure risk may show a high probability of failure through extensive sample simulations, and vice versa. The integration of these two types of information allows for mutual verification and complementarity, more accurately determining the actual risk level and effectively improving the stability prediction accuracy of double-walled steel cofferdam structures in complex environments.

[0114] Example 2:

[0115] like Figure 2 As shown, this embodiment provides an intelligent prediction system for the stability of a double-walled steel cofferdam structure. The system includes an acquisition module 901, a first processing module 902, a second processing module 903, a third processing module 904, a fourth processing module 905, and a prediction module 906, specifically including:

[0116] The acquisition module 901 is used to acquire the installation location of the sensor;

[0117] The first processing module 902 is used to arrange a sensor cluster according to the installation location of the sensors;

[0118] The second processing module 903 is used to obtain sampling information based on the data collected by the sensor cluster. The sampling information includes strain data of the double-walled steel cofferdam structure, water flow velocity data, and soil pore water pressure data.

[0119] The third processing module 904 is used to preprocess the sampling information to obtain preprocessed sampling information;

[0120] The fourth processing module 905 is used to correct the preset numerical model of the double-walled steel cofferdam based on the preprocessed sampling information to obtain the corrected numerical model.

[0121] The prediction module 906 is used to predict the stability level of the double-walled steel cofferdam structure based on the modified numerical model, and obtain the prediction result.

[0122] In one specific embodiment of this disclosure, the acquisition module further includes an acquisition unit, a first processing unit, and a second processing unit, specifically including:

[0123] The acquisition unit is used to acquire the dynamic load model and structural information of the double-walled steel cofferdam.

[0124] The first processing unit is used to determine the load of water flow on different parts of the double-walled steel cofferdam based on the dynamic load model.

[0125] The second processing unit is used to determine the installation position of the sensor based on the load of the water flow on different parts of the double-walled steel cofferdam and the structural information of the double-walled steel cofferdam.

[0126] In one specific embodiment of this disclosure, the third processing module further includes a third processing unit, a fourth processing unit, and a fifth processing unit, specifically comprising:

[0127] The third processing unit is used to process the water flow velocity data included in the sampled information using Kalman filtering to obtain the processed first information.

[0128] The fourth processing unit is used to process the soil pore water pressure data included in the sampled information using wavelet threshold denoising method to obtain the processed second information.

[0129] The fifth processing unit is used to obtain preprocessed sampling information based on the processed first information and the processed second information.

[0130] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0131] Example 3:

[0132] Corresponding to the above method embodiments, this embodiment also provides an intelligent prediction device for the stability of a double-walled steel cofferdam structure. The intelligent prediction device for the stability of a double-walled steel cofferdam structure described below and the intelligent prediction method for the stability of a double-walled steel cofferdam structure described above can be referred to in correspondence with each other.

[0133] Figure 3 This is a block diagram illustrating an intelligent prediction device 800 for the stability of a double-walled steel cofferdam structure, according to an exemplary embodiment. Figure 3 As shown, the intelligent prediction device 800 for the stability of a double-walled steel cofferdam structure may include: a processor 801 and a memory 802. The intelligent prediction device 800 may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0134] The processor 801 controls the overall operation of the intelligent prediction device 800 for the stability of the double-walled steel cofferdam structure, thereby completing all or part of the steps in the aforementioned intelligent prediction method for the stability of the double-walled steel cofferdam structure. The memory 802 stores various types of data to support the operation of the intelligent prediction device 800 for the stability of the double-walled steel cofferdam structure. This data may include, for example, instructions for any application or method operating on the intelligent prediction device 800 for the stability of the double-walled steel cofferdam structure, as well as application-related data, such as contact data, sent and received messages, images, audio, video, etc. The memory 802 can be implemented using any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 802 or transmitted via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. I / O interface 804 provides an interface between processor 801 and other interface modules, such as keyboards, mice, and buttons. These buttons can be virtual or physical. Communication component 805 is used for wired or wireless communication between the intelligent prediction device 800 for the stability of the double-walled steel cofferdam structure and other devices. Wireless communication includes, for example, Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 805 may include a Wi-Fi module, a Bluetooth module, and an NFC module.

[0135] In an exemplary embodiment, the intelligent prediction device 800 for the stability of a double-walled steel cofferdam structure can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the aforementioned intelligent prediction method for the stability of a double-walled steel cofferdam structure.

[0136] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the intelligent prediction method for the stability of a double-walled steel cofferdam structure described above. For example, the computer-readable storage medium may be the memory 802 including the program instructions described above, which may be executed by the processor 801 of the intelligent prediction device 800 for the stability of a double-walled steel cofferdam structure to complete the intelligent prediction method for the stability of a double-walled steel cofferdam structure described above.

[0137] Example 4:

[0138] Corresponding to the above method embodiments, this embodiment also provides a readable storage medium. The readable storage medium described below can be referred to in conjunction with the intelligent prediction method for the stability of a double-walled steel cofferdam structure described above.

[0139] A readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the intelligent prediction method for the stability of a double-walled steel cofferdam structure as described in the above method embodiments.

[0140] Specifically, the readable storage medium can be a USB flash drive, external hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, or any other readable storage medium capable of storing program code.

[0141] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0142] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for intelligent prediction of the stability of a double-walled steel cofferdam structure, characterized in that, include: The installation location of the sensor is obtained, and the installation location of the sensor is optimized based on the dynamic load model and the structural information of the double-walled steel cofferdam. The dynamic load model is an LSTM model. Arrange the sensor cluster according to the installation location of the sensors; Based on the data collected by the sensor cluster, sampling information is obtained, which includes strain data of the double-walled steel cofferdam structure, water flow velocity data, and soil pore water pressure data. The sampled information is preprocessed to obtain preprocessed sampled information. The preprocessing is used to suppress high-frequency vibration noise and eliminate low-frequency drift. The pre-processed sampling information is used to correct the preset numerical model of the double-walled steel cofferdam, resulting in the corrected numerical model. The stability level of the double-walled steel cofferdam structure is predicted based on the modified numerical model, and the prediction results are as follows: The physical field response data are obtained based on the modified numerical model. A deterministic safety factor is calculated based on the physical field response data. The deterministic safety factor includes an overturning safety factor, a slippage safety factor, and a seepage stability factor. A perturbation sample set is generated using the Latin hypercube sampling method. The failure probability is calculated based on the set of disturbance samples. The stability level of the double-walled steel cofferdam structure is predicted based on the deterministic safety factor and the failure probability, including: ； In the above formula, R is the overall risk value; , as well as These are the weighting coefficients corresponding to the overturning and sliding safety factors, failure probability, and seepage stability coefficient, which are calibrated based on expert experience and historical accident data. and These are the actual calculated overturning safety factors and sliding safety factors; and These are the pre-set standard values ​​for the overturning safety factor and the pre-set standard values ​​for the slippage safety factor; , , , All are adjusted indices, among which, This is used to control the nonlinear relationship between the degree to which the overturning safety factor deviates from the standard value and the overall risk value. This is used to adjust the degree to which the anti-slip safety factor deviates from the standard value and its impact on the overall risk value. Used to reflect the sensitivity of an engineering project to probabilistic failure risks. Used to control the influence of the ratio of seepage gradient to critical seepage gradient on the overall risk value; Indicates the probability of failure; Indicates the seepage stability coefficient; represents the time decay coefficient corresponding to the overturning and sliding safety factors and the failure probability, respectively; t represents the actual service time of the double-walled steel cofferdam. The process of correcting the pre-defined numerical model of the double-walled steel cofferdam based on the preprocessed sampling information includes: A state vector is constructed based on the preprocessed sampling information; Calculate the observation covariance matrix, which is used to measure the correlation and variability between state vectors; The weights of the state vector are updated based on the observed covariance matrix; The hidden parameters are determined using the particle swarm optimization algorithm, and the corrected multiphysics state parameters are obtained. The grid resolution of the preset double-walled steel cofferdam numerical model is adjusted according to the modified multiphysics state parameters to obtain the modified numerical model. The inertia weights in the particle swarm optimization algorithm are optimized as follows: ； In the above formula, and These represent the minimum and maximum values ​​of the inertia weight, respectively, used to limit the range of values ​​for the inertia weight and provide boundary conditions for the algorithm's search. This represents the fitness value of the i-th particle in the current iteration. This represents the average fitness value of particles in the current population. This represents the optimal fitness value of a particle in the current population. The specific calculation process for the overturning safety factor is as follows: ； In the above formula, W represents the self-weight of the double-walled steel cofferdam structure. This represents the horizontal distance from the line of action of the structure's self-weight to the overturning point. Indicates the frictional force of the base. This represents the horizontal distance from the line of action of the base friction force to the overturning point. and They represent water pressure and earth pressure, respectively, and H represents the vertical distance from the point of application of the water pressure to the bottom of the double-walled steel cofferdam structure; The specific calculation process for the anti-slip safety factor is as follows: ； In the above formula, The value represents the anti-slip safety factor, c represents the soil cohesion, N represents the vertical resultant force, T represents the horizontal load, and A represents the contact area between the structural base and the soil. Indicates the coefficient of friction of the substrate; The specific calculation process for the seepage stability coefficient is as follows: ； In the above formula, This represents the seepage stability coefficient. and These represent the actual hydraulic gradient and the critical hydraulic gradient, respectively.

2. The intelligent prediction method for the stability of a double-walled steel cofferdam structure according to claim 1, characterized in that, Obtain the sensor's installation location, including: Obtain the dynamic load model and structural information of the double-walled steel cofferdam; The loads on different parts of the double-walled steel cofferdam are determined based on the dynamic load model. The installation location of the sensor is determined based on the load of the water flow on different parts of the double-walled steel cofferdam and the structural information of the double-walled steel cofferdam.

3. The intelligent prediction method for the stability of a double-walled steel cofferdam structure according to claim 1, characterized in that, The sampled information is preprocessed to obtain preprocessed sampled information, including: The water flow velocity data included in the sampled information is processed using Kalman filtering to obtain the first processed information; The pore water pressure data of the soil included in the sampling information is processed using the wavelet threshold denoising method to obtain the processed second information; Based on the processed first information and the processed second information, the preprocessed sampling information is obtained.

4. The intelligent prediction method for the stability of a double-walled steel cofferdam structure according to claim 1, characterized in that, The particle swarm optimization algorithm is used to determine hidden parameters, including: Obtain the maximum and minimum values ​​of the inertia weight; The inertia weight is optimized based on the maximum and minimum values ​​of the inertia weight to obtain the optimized inertia weight. The particle position is iteratively updated based on the optimized inertia weight to obtain the optimal particle position and determine the hidden parameters.

5. A smart prediction system for the stability of a double-walled steel cofferdam structure, using any one of the smart prediction methods for the stability of a double-walled steel cofferdam structure as described in claims 1-4, characterized in that, include: The acquisition module is used to acquire the installation position of the sensor, which is optimized based on the dynamic load model and the structural information of the double-walled steel cofferdam. The first processing module is used to arrange the sensor cluster according to the installation location of the sensor; The second processing module is used to obtain sampling information based on the data collected by the sensor cluster. The sampling information includes strain data of the double-walled steel cofferdam structure, water flow velocity data, and soil pore water pressure data. The third processing module is used to preprocess the sampling information to obtain preprocessed sampling information. The preprocessing is used to suppress high-frequency vibration noise and eliminate low-frequency drift. The fourth processing module is used to correct the preset numerical model of the double-walled steel cofferdam based on the preprocessed sampling information to obtain the corrected numerical model. The prediction module is used to predict the stability level of the double-walled steel cofferdam structure based on the modified numerical model, and obtain the prediction results.

6. The intelligent prediction system for the stability of a double-walled steel cofferdam structure according to claim 5, characterized in that, The acquisition module includes: The acquisition unit is used to acquire the dynamic load model and structural information of the double-walled steel cofferdam. The first processing unit is used to determine the load of water flow on different parts of the double-walled steel cofferdam based on the dynamic load model. The second processing unit is used to determine the installation position of the sensor based on the load of the water flow on different parts of the double-walled steel cofferdam and the structural information of the double-walled steel cofferdam.

7. The intelligent prediction system for the stability of a double-walled steel cofferdam structure according to claim 5, characterized in that, The third processing module includes: The third processing unit is used to process the water flow velocity data included in the sampled information using Kalman filtering to obtain the processed first information. The fourth processing unit is used to process the soil pore water pressure data included in the sampled information using wavelet threshold denoising method to obtain the processed second information. The fifth processing unit is used to obtain preprocessed sampling information based on the processed first information and the processed second information.

Images