Exterior river retaining wall displacement monitoring system
Through multi-sensor data fusion and nonlinear modeling, combined with L-BFGS-B optimization and weighted Kalman filtering algorithms, the sensor layout is dynamically optimized and the deformation field model is updated. This solves the problem of accurate and real-time monitoring of the outer river retaining wall monitoring system under complex working conditions, and achieves high-precision and stable monitoring effects.
Patent Information
- Application Number
- CN202510952125.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-10
- Publication Date
- 2025-09-12
AI Technical Summary
The existing monitoring system for retaining walls in Zhongwai River relies on a single sensor and has simple data processing. It cannot achieve accurate and real-time deformation assessment under complex working conditions. The sensor layout is static and lacks dynamic optimization. The deformation field model cannot be updated in a timely manner, resulting in a decrease in monitoring accuracy.
By adopting multi-sensor data fusion and nonlinear elastic mechanics modeling, combined with the L-BFGS-B optimization algorithm and weighted Kalman filter algorithm, the sensor layout is dynamically optimized and the deformation field model is updated to achieve high-precision and real-time monitoring.
It improves the accuracy and reliability of the monitoring system, solves the problems of redundant sensor layout and incomplete coverage, ensures data accuracy and stability in complex environments, and realizes real-time and accurate monitoring of the deformation of the outer river retaining wall.
Smart Images
Figure CN120627993A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of water conservancy project safety monitoring, in particular to an outer river retaining wall displacement monitoring system. Background Art
[0002] As a crucial protective structure, retaining walls are widely used in water conservancy projects. Over time, these walls are susceptible to external factors, such as water impact and soil pressure. These factors can cause the structure to shift, tilt, or twist, compromising overall safety. Therefore, real-time monitoring of the deformation of retaining walls is crucial. Traditional monitoring methods have numerous limitations and cannot meet the precision and real-time requirements required under complex working conditions.
[0003] Existing monitoring systems for external river retaining walls typically rely on a single type of sensor to capture deformation data in a specific direction. While these sensors can provide a certain degree of insight into the forces acting on the retaining wall, they often struggle to comprehensively and accurately assess the overall deformation of the wall when subjected to multiple external forces. Furthermore, some technologies employ simple linear models for calculation. While these models can provide some guidance, they lack sufficient flexibility and accuracy, and are often unsatisfactory when dealing with complex and nonlinear deformations.
[0004] Traditional monitoring methods process data in a relatively simplistic manner, especially when fusing multi-sensor data, often employing only an average approach. This approach ignores differences in measurement accuracy between different sensors and can easily lead to deviations in the monitoring data. Furthermore, most existing sensor layouts are static and lack dynamic optimization based on real-time data. Consequently, insufficient coverage or blind spots may exist during monitoring, preventing the monitored area from fully reflecting the actual deformation. More importantly, deformation field models are mostly static, making them difficult to adapt to changing operating conditions and unable to be updated and adjusted in a timely manner, resulting in a continuous decline in accuracy during long-term monitoring. Summary of the Invention
[0005] In view of the shortcomings of the existing technology, the present invention provides an outer river retaining wall displacement monitoring system, which solves the problems of static sensor layout, inaccurate data fusion and inability to dynamically update the deformation field model in the existing technology.
[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A displacement monitoring system for an outer river retaining wall, comprising: A deformation field modeling module is used to establish a nonlinear elastic mechanics model based on the physical characteristics and stress conditions of the outer river retaining wall. The nonlinear elastic mechanics model can simulate the displacement, tilt and distortion of the retaining wall under the influence of various external factors; A data acquisition module connected to the deformation field modeling module, wherein the data acquisition module is used to deploy three sensors, and the three sensors are used to collect original data of displacement, tilt and distortion of each monitoring point of the outer river retaining wall; a data processing module, connected to the data acquisition module and the deformation field modeling module, respectively, the data processing module receives the raw data from the data acquisition module and calculates the deformation field of each monitoring point based on the deformation field model, and optimizes the sensor position and data acquisition scheme; The optimization module is connected to the data processing module. The optimization module uses the L-BFGS-B optimization algorithm to calculate the error between the deformation field of each monitoring point based on the original data and the deformation field model, optimizes the sensor layout and minimizes the error; the data fusion module is connected to the optimization module. The data fusion module uses the weighted Kalman filter algorithm to fuse the original data collected by the three sensors.
[0007] Preferably, the deformation field modeling module includes: Nonlinear elastic mechanics modeling unit, used to establish the nonlinear deformation model of the outer river retaining wall under the action of external factors; The deformation field calculation unit is used to calculate the displacement, tilt and distortion of each monitoring point based on the nonlinear elastic mechanics model; the model update unit is used to dynamically update the deformation field model in combination with the actual monitoring data after collecting new data in real time.
[0008] Preferably, the deformation field calculation unit calculates the deformation field according to the following nonlinear elastic mechanics equation: Where F is the total force; σ(u) is the stress tensor; u is the displacement vector; is the displacement gradient; F ext is the external force; dV is the volume element.
[0009] Preferably, the data acquisition module includes: Horizontal displacement sensor, vertical displacement sensor and tilt sensor, the three sensors are arranged at three key monitoring points of the outer river retaining wall, and are used to collect displacement data in the horizontal direction, vertical direction and tilt direction respectively; The data transmission unit is used to transmit the raw data collected by the sensor to the data processing module in real time through the wireless network.
[0010] Preferably, the data transmission unit includes: A wireless communication module, used to transmit the raw data collected by the sensor to the data processing module via a wireless network, wherein the wireless communication module supports a standard wireless communication protocol; A data cache unit is used to cache sensor data when the wireless signal is unstable or the network is interrupted, and automatically upload the cached data when the signal is restored; The power management unit is used to provide a stable power supply to ensure that the data transmission unit can work normally under various environmental conditions.
[0011] Preferably, the data processing module includes: A data receiving unit, used for receiving raw data from the data acquisition module; A data analysis unit calculates the displacement, tilt and deformation parameters of each monitoring point based on the deformation field model and performs error analysis on the collected raw data; The optimization suggestion unit proposes optimization of sensor layout based on data analysis results.
[0012] Preferably, the data analysis unit includes: The data fitting unit is used to fit the actual collected data of each monitoring point with the expected data calculated based on the deformation field model to obtain a deformation field with the minimum error; Anomaly detection unit, used to identify outliers in the data and automatically judge and mark abnormal data by comparing the monitored data with historical data or theoretical models; The data filtering unit is used to perform noise suppression processing on the collected raw data and filter out high-frequency noise and interference signals.
[0013] Preferably, the optimization module includes: L-BFGS-B optimization algorithm unit, used to adjust the sensor layout based on the error between the original data and the deformation field calculated by the deformation field model, and minimize the error; The optimization strategy unit is used to adjust the parameters of the monitoring system according to the optimization results of the L-BFGS-B algorithm, including the location of sensors, data collection frequency and processing method.
[0014] Preferably, the L-BFGS-B optimization algorithm unit further includes: The multi-objective optimization unit is used to simultaneously consider multiple objective functions during the optimization process. The multiple objective functions include monitoring accuracy, sensor resource usage, and system response time. The optimization objectives can be expressed by the following objective functions: Where u i is the actual data of the monitoring point; is the model calculation result; P j is the deviation of sensor arrangement; T k is the response time of the sensor; λ1 and λ2 are weight coefficients; min means minimization.
[0015] Preferably, the data fusion module includes: The Kalman filter unit is used to fuse the raw data from multiple sensors through the weighted Kalman filter algorithm. Specifically, the weighted Kalman filter formula is used: Where, is the estimated value of Kalman filter; K k is the Kalman gain; z k is the measurement data from the sensor; is the measurement matrix; The data correction unit is used to correct the sensor data according to the output of the Kalman filter, eliminate redundant data and reduce measurement errors.
[0016] The present invention provides a system for monitoring the displacement of an outer river retaining wall. It has the following beneficial effects: 1. This invention utilizes a technical solution that combines a data acquisition module with a deformation field modeling module, achieving efficient, real-time collection of deformation data from the outer river retaining wall and accurate calculations based on the deformation field model. Compared to existing solutions that rely solely on a single sensor or simple model, this invention addresses the inaccuracy and lack of real-time performance associated with a single data source through multi-sensor data fusion and complex nonlinear modeling, thereby improving monitoring accuracy and reliability.
[0017] 2. This invention optimizes sensor layout using the L-BFGS-B optimization algorithm, ensuring high monitoring accuracy while minimizing sensor redundancy and cost. Compared to existing solutions that use fixed or random layouts, this invention dynamically optimizes sensor positions, effectively solving the problems of incomplete coverage and resource waste in the monitoring area, providing a more flexible and economical solution for long-term, large-scale monitoring.
[0018] 3. This invention uses a weighted Kalman filter algorithm to fuse data collected by multiple sensors, improving data accuracy and anti-interference capabilities. Unlike existing solutions that use simple averaging or unweighted fusion, this invention performs weighted processing based on the data quality and reliability of each sensor, ensuring the accuracy and stability of monitoring data in complex environments and effectively overcoming the impact of noise interference and data loss.
[0019] 4. This invention utilizes dynamic deformation field model updates to enable the monitoring system to accurately reflect the real-time deformation status of the outer river retaining wall. Compared to existing solutions that rely on static models, this invention uses real-time data to drive the update of the deformation field, enabling the monitoring system to promptly respond to changes in the external environment. This solves the problems of model lag and reduced accuracy during long-term monitoring, providing more accurate and dynamic monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 This is a system architecture diagram of the present invention; Figure 2 This is a framework diagram of the deformation field modeling module of the present invention; Figure 3 This is a data acquisition module framework diagram of the present invention; Figure 4 This is a data processing module framework diagram of the present invention; Figure 5 This is the optimization module framework diagram of the present invention; Figure 6 This is a framework diagram of the data fusion module of the present invention. DETAILED DESCRIPTION
[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.
[0022] Please see the attached Figure 1 -Attached Figure 6 The embodiment of the present invention provides a system for monitoring displacement of an outer river retaining wall, comprising: A deformation field modeling module is used to establish a nonlinear elastic mechanics model based on the physical characteristics and stress conditions of the outer river retaining wall. The nonlinear elastic mechanics model can simulate the displacement, tilt and distortion of the retaining wall under the influence of various external factors; The deformation field modeling module is designed to establish a nonlinear elastic mechanics model based on the physical properties and stress conditions of the outer river retaining wall. This model not only simulates the displacement, tilt, and distortion of the outer river retaining wall under common external factors, but also responds in real time to complex environmental conditions such as water flow and soil pressure. This modeling module provides accurate basic data for subsequent data processing and optimization, ensuring the accuracy and stability of monitoring results.
[0023] Alternatively, the deformation field modeling module establishes a nonlinear elastic mechanics model by considering the material properties (e.g., elastic modulus, Poisson's ratio, etc.), geometry (e.g., height, width, thickness, etc.), and external forces (e.g., water flow, soil pressure, earthquakes, etc.) of the outer river retaining wall. Specifically, this embodiment uses a combination of the finite element method and the nonlinear elastic mechanics model to model the deformation field of the outer river retaining wall.
[0024] Typically, this module dynamically updates the deformation field model based on the physical properties of the retaining wall and actual monitoring data, thereby obtaining real-time deformation information for each monitoring point. By accurately simulating the displacement, tilt, and distortion of the retaining wall under the influence of various external factors, it provides accurate input data for other modules of the monitoring system, further optimizing data processing and sensor placement decisions.
[0025] Nonlinear Elasticity Modeling: This unit describes the deformation behavior of the retaining wall by constructing a nonlinear elasticity model based on the physical properties of the retaining wall and the external forces acting on it. This process requires consideration of the nonlinear mechanical properties of the retaining wall material and the complex relationship between deformation and external forces. The following basic nonlinear elasticity equations are used in this stage: Where F is the total force; σ(u) is the stress tensor; u is the displacement vector; is the displacement gradient; F ext is the external force; dV is the volume element.
[0026] In the specific calculation, It represents the gradient of the displacement field and reflects the rate of change of displacement. ext It may include various factors such as water flow pressure, soil forces, temperature changes, etc.
[0027] Deformation Field Calculation Unit: Based on an established nonlinear elastic mechanics model, this unit performs numerical calculations using methods such as finite element analysis (FEA) to determine deformation parameters such as displacement, tilt, and twist at each monitoring point. By inputting the retaining wall's physical properties and stress conditions into the calculation model, the Deformation Field Calculation Unit can quickly and accurately calculate the wall's deformation. The unit's calculation results provide a foundation for subsequent data processing and optimization modules.
[0028] Model Update Unit: As the external environment (such as water level changes, soil pressure changes, etc.) and sensor data continue to change, the model update unit dynamically updates the original model based on real-time monitoring data and model calculation results. The updated model will better fit the actual deformation situation, thereby improving the accuracy and real-time performance of deformation field prediction.
[0029] In some embodiments, the nonlinear elastic mechanics model not only considers lateral and longitudinal displacement but also incorporates factors such as the wall's tilt and twist, performing multi-dimensional calculations to comprehensively assess the wall's stress and deformation. For example, when subjected to the combined effects of irregular soil pressure and uneven water flow, the model needs to simultaneously analyze the entire wall's deformation to ensure accurate monitoring data.
[0030] Through the deformation field modeling module described in this embodiment, the system can accurately simulate the external river retaining wall under complex external environments. The nonlinear elastic mechanics model can reflect the actual deformation of the external river retaining wall under various forces, particularly its displacement, tilt, and distortion under dynamic conditions such as water flow and soil pressure fluctuations. This enables the monitoring system of the present invention to accurately track the deformation of the retaining wall in real time and promptly identify potential safety hazards.
[0031] The technical solution of this embodiment not only achieves high accuracy, but also, through the integration of finite element analysis and dynamic model updating, provides the system with excellent adaptability and stability. The presence of the model update unit ensures continuous optimization of the deformation field model in practical applications, enabling the system to maintain high accuracy even during long-term operation. By accurately analyzing and simulating the stress state of the retaining wall, this invention can provide reliable data support for subsequent optimization, data processing, and decision-making.
[0032] The data acquisition module is connected to the deformation field modeling module. The data acquisition module is used to deploy three sensors, and the three sensors are used to collect the original data of displacement, tilt and distortion of each monitoring point of the outer river retaining wall. The data acquisition module is an important part of the outer river retaining wall displacement monitoring system. It is responsible for real-time collection of the original data of each monitoring point of the outer river retaining wall and transmits it to the subsequent data processing module. This module is closely connected with the deformation field modeling module and can provide the necessary original input data for the calculation of the deformation field model. Specifically, the data acquisition module is used to collect the displacement, tilt and distortion data of the outer river retaining wall in three directions by deploying three different types of sensors. These data are the basic information relied on by the deformation field modeling module and the subsequent optimization and data processing modules.
[0033] In general, the data acquisition module, through real-time data collection from various monitoring points, accurately reflects the deformation of the retaining wall during stress loading and provides important input for accurate calculation of the deformation field model. The sensor configuration of the data acquisition module ensures comprehensive monitoring of changes in different directions of the outer river retaining wall.
[0034] The data acquisition module is equipped with three sensors to collect displacement, tilt and distortion data at each monitoring point of the outer river retaining wall. These sensors mainly include: Horizontal displacement sensor, used to monitor the horizontal displacement of the retaining wall.
[0035] Vertical displacement sensor, used to monitor the vertical displacement of the retaining wall.
[0036] Inclination sensor, used to monitor the inclination angle of the retaining wall.
[0037] Horizontal displacement sensor: This sensor measures the horizontal displacement of a retaining wall, providing deformation data when subjected to horizontal external forces. Typically, a horizontal displacement sensor uses a laser displacement meter or capacitive displacement sensor. Based on the measurement principle, it calculates deformation by detecting the relative horizontal position change of the target object in real time.
[0038] Vertical displacement sensor: This sensor collects vertical displacement data of the retaining wall. By measuring the displacement of the retaining wall when subjected to vertical forces such as water flow and soil pressure, the vertical displacement sensor can provide real-time vertical displacement changes of the retaining wall. Data collection is performed using a common displacement meter or fiber optic sensor.
[0039] Inclination sensors are used to measure the tilt of retaining walls. By detecting changes in the retaining wall's angle relative to the vertical in real time, they can accurately capture the wall's tilt under external disturbances. These sensors typically utilize precision electronic inclinometers or accelerometers, enabling real-time monitoring of the retaining wall's tilt within a very low accuracy range.
[0040] The data acquisition module also includes a data transmission unit, which transmits the collected raw data to the data processing module in real time via a wireless network. This unit utilizes efficient communication protocols to ensure real-time and stable data transmission. The wireless communication module supports common communication protocols such as Wi-Fi, Zigbee, and LoRa, ensuring fast data transmission in various network environments.
[0041] In some embodiments, the data transmission unit is also equipped with a data cache unit. When the network signal is unstable or interrupted, the cache unit can temporarily store sensor data and automatically upload the cached data when the signal is restored, preventing data loss. The data transmission unit also includes a power management unit to ensure stable data collection and transmission during long-term operation.
[0042] To ensure the accuracy of the monitoring system, the sensors in the data acquisition module are highly precise and stable. In practice, these sensors accurately measure the displacement and deformation of the outer river retaining wall under the influence of various external factors. Furthermore, the sensor layout strategy has been carefully designed to ensure coverage of key areas of the retaining wall, thereby capturing any deformation anomalies in real time.
[0043] The data acquisition module is connected to the aforementioned deformation field modeling module through real-time data transmission. The deformation field modeling module receives the raw displacement, tilt, and distortion data provided by the data acquisition module and uses this data as input to calculate the deformation field of the retaining wall. Based on this real-time data, the deformation field modeling module updates various parameters in the model to ensure more accurate deformation field calculations.
[0044] Specifically, the displacement, tilt, and distortion data collected by the data acquisition module at each monitoring point during the monitoring process provide the necessary input information for the deformation field model, enabling it to better simulate the deformation state of the retaining wall and providing a basis for subsequent data processing and optimization modules. The model update unit uses this data to update the deformation field model in real time, further improving the adaptability and accuracy of the monitoring system.
[0045] The data acquisition module in this embodiment, through precise sensor configuration and efficient wireless data transmission, ensures that raw data on displacement, tilt, and distortion at each monitoring point on the outer river retaining wall is accurately transmitted to the data processing module in real time. This module ensures that the monitoring system can accurately simulate and analyze the retaining wall's deformation based on real-world monitoring data.
[0046] Through the data acquisition module, the monitoring system can cover multiple key locations of the outer river retaining wall, comprehensively monitoring its deformation under external forces. This not only improves the coverage of monitoring data but also provides reliable raw data for subsequent data processing and deformation field modeling.
[0047] a data processing module, connected to the data acquisition module and the deformation field modeling module, respectively, the data processing module receives the raw data from the data acquisition module and calculates the deformation field of each monitoring point based on the deformation field model, and optimizes the sensor position and data acquisition scheme; After the aforementioned data acquisition and deformation field modeling modules collaborate to acquire basic monitoring information and conduct physical modeling, the data processing module, acting as the connecting link within the entire system, performs data analysis, modeling calculations, and optimization feedback. This module receives raw monitoring data from the data acquisition module and, relying on the nonlinear elastic mechanics model constructed by the deformation field modeling module, deduces and restores the overall deformation field structure of the outer river retaining wall.
[0048] Typically, the data processing module normalizes the raw data, extracts features, and identifies anomalies. It then combines existing model parameters to calculate deformation and analyze the rationality of the current sensor spatial distribution. Furthermore, the processing results can be used to guide subsequent updates and optimizations of acquisition strategies, thus forming a data-driven closed-loop optimization mechanism.
[0049] In this embodiment, the data processing module is connected to the data acquisition module and the deformation field modeling module respectively, and its main functions include data reception, model-driven deformation field reconstruction, and sensor position optimization and acquisition strategy adjustment.
[0050] As an option, the data processing module includes a data preprocessing unit, a model fitting unit, an error analysis unit and an optimization feedback unit.
[0051] In one possible implementation, the model fitting unit calculates the deformation field based on the aforementioned nonlinear elastic model. Unlike traditional displacement field recovery methods, this unit uses a multi-point coupling solution strategy to restore the local displacement state of each monitoring point through the full-field stress response. This process relies on the following deformation recovery formula: Where u i is the actual data of the monitoring point; is the point-to-point stiffness coupling matrix obtained during the modeling process; f j is the equivalent external force at the jth monitoring point; Represents the nonlinear solution process of recovering the displacement according to the inverse function of the modeling formula.
[0052] The above calculation process not only reflects the nonlinear coupling relationship between force and deformation, but also reflects the practical application of high-order tensor fields in deformation mapping.
[0053] In the data processing module, the error analysis unit is used to evaluate the error distribution between the actual collected data and the theoretical model prediction results, and further used to evaluate the effectiveness of the sensor deployment. The error function can take the following form: Where, E is the overall error index; N is the number of sensor monitoring points; is the actual deformation vector collected for the i-th point; For modeling predictions.
[0054] When the local area error is significantly higher than the global average error, the area is marked as a candidate area for layout optimization and used for subsequent layout adjustment strategy generation.
[0055] Based on the error analysis results, the optimization feedback unit automatically generates a new sensor layout plan through heuristic search or gradient guidance mechanism. In one possible implementation, the following layout adjustment index function is used: Where ΔP represents the placement adjustment vector; p is the current sensor spatial coordinate; E is the overall error index; C is the system communication or placement cost; and λ is the adjustment weight coefficient, balancing accuracy and cost.
[0056] This strategy enables the data processing module to not only have the functions of passive reception and solution, but also the ability to actively intervene for future optimization and adjustment, forming a three-in-one intelligent processing architecture of "data-modeling-feedback".
[0057] In some embodiments, to reduce the computational burden of the system, the data processing module introduces approximate solution strategies or low-rank modeling methods, reconstructing key displacement patterns through singular value decomposition (SVD) or sparse coding to improve computational efficiency. Such methods have strong adaptability in large-scale structural monitoring.
[0058] An optimization module is connected to the data processing module. The optimization module uses the L-BFGS-B optimization algorithm to calculate the error between the deformation field of each monitoring point based on the original data and the deformation field model, adjusts the sensor layout and minimizes the error; The optimization module is a key component of the outer river retaining wall displacement monitoring system. Its primary function is to optimize the sensor layout and minimize errors using the L-BFGS-B optimization algorithm, based on the raw data and deformation field model calculations provided by the data processing module. The optimization module is connected to the data processing module and dynamically adjusts the sensor layout based on error information provided by the data analysis unit, ensuring high accuracy and efficiency of the monitoring system.
[0059] Typically, the optimization module uses the discrepancy between raw data and the deformation field calculated using the deformation field model to perform error analysis. Using the L-BFGS-B algorithm, the system adjusts for discrepancies between the actual deformation at monitoring points and the model's calculated results, optimizing sensor placement and data collection strategies. This effectively improves monitoring accuracy, reduces redundant placement, and ensures the stability and reliability of the monitoring system.
[0060] In this embodiment, the optimization module receives the raw data and deformation field model calculation results output by the data processing module. The core goal of the error calculation is to evaluate the difference between the calculated displacement, tilt, and distortion values of each monitoring point and the actual collected values. To this end, the following error formula can be used for quantification: Where E is the overall error index; is the actual measurement data of the i-th monitoring point; is the displacement data calculated based on the deformation field model; N is the number of sensor monitoring points.
[0061] This error measurement standard helps the optimization module determine the deviation between the deformation field model and the collected data, thereby providing a basis for subsequent optimization adjustments.
[0062] The optimization module minimizes the error using the L-BFGS-B optimization algorithm. L-BFGS-B is a finite-memory Newton method commonly used for large-scale optimization problems. It efficiently optimizes within limited memory. Specifically, at each optimization step, the L-BFGS-B algorithm guides the optimization process by calculating the gradient of the objective function, thereby gradually updating the sensor placement plan.
[0063] During implementation, multiple objective functions include monitoring accuracy, sensor resource usage, and system response time. The optimization goal can be expressed as the following objective function: Where u i is the actual data of the monitoring point; is the model calculation result; P j is the deviation of sensor arrangement; T k is the response time of the sensor; λ1 and λ2 are weight coefficients; min means minimization.
[0064] After the optimization module outputs the optimization results, the sensor layout adjustments are fed back to the data processing module in real time. Based on the optimized sensor layout, the data processing module further performs data processing and analysis tasks to ensure that the data collected under the new sensor layout accurately reflects the displacement, tilt, and distortion of the outer river retaining wall.
[0065] Alternatively, the optimization module can dynamically adjust its optimization strategy based on different scenario requirements. For example, in some areas, it might prioritize improving monitoring accuracy, while in other areas, it might prioritize monitoring costs and efficient use of sensor resources. By adjusting the optimization algorithm, the monitoring system can flexibly adapt to different deployment requirements.
[0066] In some embodiments, the optimization module can also be expanded to a multi-objective optimization problem, considering multiple objective functions, such as monitoring accuracy, sensor resource utilization, and system response time. In this case, the optimization process is not only about minimizing the error, but also requires balancing multiple objectives.
[0067] A data fusion module is connected to the optimization module, and the data fusion module adopts a weighted Kalman filter algorithm to fuse the raw data collected by the three sensors; The data fusion module, connected to the optimization module, is responsible for fusing the raw data collected by the three sensors to optimize data accuracy and reliability. The data fusion module uses a weighted Kalman filter algorithm to process the data from multiple sensors, thereby improving the accuracy of the final monitoring results and ensuring the monitoring system has high robustness and accuracy in practical applications.
[0068] Typically, the data acquisition module deploys horizontal displacement sensors, vertical displacement sensors, and inclination sensors to collect data on the displacement, tilt, and distortion of the outer river retaining wall in three directions. This raw data can be affected by various external factors, such as signal noise and sensor errors. Therefore, it requires precise processing by the data fusion module to improve data reliability and accuracy.
[0069] Specifically, the data fusion module uses a weighted Kalman filter algorithm and consists of two main components: state estimation and covariance updating. In state estimation, the data fusion module adjusts the estimated values of state variables based on the discrepancies between the system's predicted model and the actual measured data. Covariance updating further optimizes the state estimation by calculating the difference between the new estimated values and the original measured values.
[0070] In this embodiment, the Kalman filter unit is used to fuse the raw data from multiple sensors using a weighted Kalman filter algorithm, specifically using the weighted Kalman filter formula: Where, is the estimated value of Kalman filter; K k is the Kalman gain; z k is the measurement data from the sensor; is the measurement matrix.
[0071] In this algorithm, the Kalman gain K k The calculation formula is: Where R k is the measurement noise covariance matrix, which represents the uncertainty of the measurement data; K k is the Kalman gain; represents the transpose of the measurement matrix; P k-1 is the prior error covariance matrix; is the measurement matrix; R k is the measurement noise covariance matrix.
[0072] During the data fusion process, the Kalman filter weights the data from each sensor according to its weight and combines the data from multiple sensors into an optimal estimate. Specifically, each sensor's measurement is compared with the system's predicted value, and the Kalman gain adjusts the weights to more accurately estimate the system's current state. The data fusion module takes into account the noise and uncertainty of the sensor measurement data to calculate more accurate displacement, tilt, and distortion parameters.
[0073] In one possible implementation, the data fusion module not only considers the error of a single sensor, but also dynamically adjusts the weighting coefficients of each sensor based on factors such as sensor quality, location, and historical performance. Specifically, the following weighted Kalman filter formula can be used: Where, is the estimated value of Kalman filter; w i is the weight coefficient of the i-th sensor; The state estimate calculated for the i-th sensor.
[0074] By applying a weighted Kalman filter, the data fusion module can effectively reduce noise in the raw data, improving data accuracy and stability. In some embodiments, the data fusion module also incorporates other algorithms, such as adaptive filtering or particle filtering, to further optimize the data processing process and ensure accurate fusion of sensor data under various environmental conditions.
[0075] Alternatively, the data fusion module can perform real-time data processing within the optimized sensor layout to dynamically adjust monitoring accuracy. This approach ensures the system maintains high accuracy under changing environmental conditions and adapts to different monitoring tasks.
[0076] The data after Kalman filtering and weighting is fed back to the data processing module as optimization results. The data processing module further uses this optimized data to calculate the deformation field and optimize the sensor layout. Ultimately, the output of the data fusion module not only improves the accuracy of the data from individual sensors but also enhances the robustness of the entire monitoring system, enabling the system to effectively cope with potential issues such as environmental noise and data loss.
[0077] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.
Claims
1. An outer river retaining wall displacement monitoring system, characterized in that: include: A deformation field modeling module is used to establish a nonlinear elastic mechanics model based on the physical characteristics and stress conditions of the outer river retaining wall. The nonlinear elastic mechanics model can simulate the displacement, tilt and distortion of the retaining wall under the influence of various external factors; A data acquisition module connected to the deformation field modeling module, wherein the data acquisition module is used to deploy three sensors, and the three sensors are used to collect original data of displacement, tilt and distortion of each monitoring point of the outer river retaining wall; a data processing module, connected to the data acquisition module and the deformation field modeling module, respectively, the data processing module receives the raw data from the data acquisition module and calculates the deformation field of each monitoring point based on the deformation field model, and optimizes the sensor position and data acquisition scheme; The optimization module is connected to the data processing module. The optimization module uses the L-BFGS-B optimization algorithm to calculate the error between the deformation field of each monitoring point based on the original data and the deformation field model, optimizes the sensor layout and minimizes the error; the data fusion module is connected to the optimization module. The data fusion module uses the weighted Kalman filter algorithm to fuse the original data collected by the three sensors.
2. The outer river retaining wall displacement monitoring system according to claim 1 is characterized in that: The deformation field modeling module includes: Nonlinear elastic mechanics modeling unit, used to establish the nonlinear deformation model of the outer river retaining wall under the action of external factors; The deformation field calculation unit is used to calculate the displacement, tilt and distortion of each monitoring point based on the nonlinear elastic mechanics model; the model update unit is used to dynamically update the deformation field model in combination with the actual monitoring data after collecting new data in real time.
3. The outer river retaining wall displacement monitoring system according to claim 2, characterized in that: The deformation field calculation unit calculates the deformation field according to the following nonlinear elastic mechanics equation: Where F is the total force; σ(u) is the stress tensor; u is the displacement vector; is the displacement gradient; F ext is the external force; dV is the volume element.
4. The outer river retaining wall displacement monitoring system according to claim 1, characterized in that: The data acquisition module includes: Horizontal displacement sensor, vertical displacement sensor and tilt sensor, the three sensors are arranged at three key monitoring points of the outer river retaining wall, and are used to collect displacement data in the horizontal direction, vertical direction and tilt direction respectively; The data transmission unit is used to transmit the raw data collected by the sensor to the data processing module in real time through the wireless network.
5. The outer river retaining wall displacement monitoring system according to claim 4 is characterized in that: The data transmission unit includes: A wireless communication module, used to transmit the raw data collected by the sensor to the data processing module via a wireless network, wherein the wireless communication module supports a standard wireless communication protocol; A data cache unit is used to cache sensor data when the wireless signal is unstable or the network is interrupted, and automatically upload the cached data when the signal is restored; The power management unit is used to provide a stable power supply to ensure that the data transmission unit can work normally under various environmental conditions.
6. The outer river retaining wall displacement monitoring system according to claim 1, characterized in that: The data processing module includes: A data receiving unit, used for receiving raw data from the data acquisition module; A data analysis unit calculates the displacement, tilt and deformation parameters of each monitoring point based on the deformation field model and performs error analysis on the collected raw data; The optimization suggestion unit proposes optimization of sensor layout based on data analysis results.
7. The outer river retaining wall displacement monitoring system according to claim 6, characterized in that: The data analysis unit includes: The data fitting unit is used to fit the actual collected data of each monitoring point with the expected data calculated based on the deformation field model to obtain a deformation field with the minimum error; Anomaly detection unit, used to identify outliers in the data and automatically judge and mark abnormal data by comparing the monitored data with historical data or theoretical models; The data filtering unit is used to perform noise suppression processing on the collected raw data and filter out high-frequency noise and interference signals.
8. The outer river retaining wall displacement monitoring system according to claim 1, characterized in that: The optimization module includes: an L-BFGS-B optimization algorithm unit for adjusting the sensor arrangement based on the error between the original data and the deformation field calculated by the deformation field model, and minimizing the error; The optimization strategy unit is used to adjust the parameters of the monitoring system according to the optimization results of the L-BFGS-B algorithm, including the location of sensors, data collection frequency and processing method.
9. The outer river retaining wall displacement monitoring system according to claim 8, characterized in that: The L-BFGS-B optimization algorithm unit further includes: The multi-objective optimization unit is used to simultaneously consider multiple objective functions during the optimization process. The multiple objective functions include monitoring accuracy, sensor resource usage, and system response time. The optimization objectives can be expressed by the following objective functions: Where u i is the actual data of the monitoring point; is the model calculation result; P j is the deviation of sensor arrangement; T k is the response time of the sensor; λ1 and λ2 are weight coefficients; min means minimization.
10. The outer river retaining wall displacement monitoring system according to claim 1, characterized in that: The data fusion module includes: The Kalman filter unit is used to fuse the raw data from multiple sensors through the weighted Kalman filter algorithm. Specifically, the weighted Kalman filter formula is used: Where, is the estimated value of Kalman filter; K k is the Kalman gain; z k is the measurement data from the sensor; is the measurement matrix; The data correction unit is used to correct the sensor data according to the output of the Kalman filter, eliminate redundant data and reduce measurement errors.