A rotating bridge posture monitoring system based on dynamic feedback control

Abstract

The present application discloses a rotating bridge posture monitoring system based on dynamic feedback control, which belongs to the field of bridge construction technology. It includes the following modules: a multi-source sensor integration module, which builds an all-round posture monitoring network to achieve high-precision real-time monitoring of key parts of the bridge; an edge computing and data fusion module, which realizes real-time on-site processing and fusion of sensor data, effectively reduces data transmission delay and improves monitoring stability; a multi-physics field coupling modeling module, which builds a multi-physics field coupling model including structural mechanics, temperature field, contact friction and fluid action, and realizes accurate prediction of the posture change of the rotating bridge; an adaptive control strategy module, which realizes high-precision adaptive control of the posture of the rotating bridge; a graded warning and collaborative control module, which automatically matches the corresponding control strategy, realizes risk graded response and intelligent collaborative control; a digital twin interactive verification module, which builds a high-precision digital twin model of the rotating bridge.

Description

Technical Field

[0001] The present application relates to the technical field of bridge construction, and more specifically, to a rotation bridge posture monitoring system based on dynamic feedback control. Background Art

[0002] As an important engineering structure, rotating bridges are widely used in ports, waterways, urban transportation and other fields. They have the unique function of rotating and opening at a specific time to pass large ships or vehicles. With the improvement of modern transportation needs and the continuous increase in the scale of bridges, the attitude control and monitoring technology of rotating bridges has become increasingly important. At present, traditional rotating bridge attitude control systems generally rely on rigid adjustment and manual intervention of mechanical equipment, which is difficult to cope with complex external environmental changes and internal working conditions, resulting in poor system stability and a high frequency of accidents. In addition, the bridge faces complex mechanical effects during the rotation process, such as structural stress, temperature changes, wind loads and other factors, which put higher requirements on the stability and accuracy of the bridge.

[0003] Existing bridge posture monitoring systems mostly rely on single sensor data for analysis and lack effective data fusion and real-time processing mechanisms, resulting in data transmission delays and difficulty in ensuring monitoring accuracy and real-time performance. Although some studies have proposed a variety of optimization solutions, such as the introduction of intelligent control and prediction models, due to the complexity of rotating bridges and changing environmental factors, existing technologies still find it difficult to achieve accurate, dynamic, and adaptive posture control and monitoring.

[0004] In order to ensure the safety and accuracy of the rotating bridge during operation, how to combine sensor data, edge computing and advanced control algorithms for real-time monitoring and feedback control in a complex external environment has become a technical problem that needs to be solved urgently. Summary of the invention

[0005] In order to overcome a series of defects in the prior art, the purpose of this application is to provide a rotating bridge posture monitoring system based on dynamic feedback control, including the following modules, in view of the above problems.

[0006] The multi-source sensor integration module integrates multiple types of sensor devices, builds an all-round posture monitoring network, and realizes high-precision real-time monitoring of key parts of the bridge.

[0007] The edge computing and data fusion module uses the edge computing nodes deployed on-site and the multi-scale Kalman filtering algorithm to realize real-time on-site processing and fusion of sensor data, effectively reducing data transmission delays and improving monitoring stability.

[0008] The multi-physics coupling modeling module builds a multi-physics coupling model including structural mechanics, temperature field, contact friction and fluid action to achieve accurate prediction of the posture changes of the rotating bridge.

[0009] The adaptive control strategy module dynamically optimizes control parameters based on real-time monitoring data, combining model predictive control, parameter adaptation and reinforcement learning technology to achieve high-precision adaptive control of the rotating bridge posture.

[0010] The hierarchical warning and collaborative control module automatically matches the corresponding control strategies by establishing a multi-level warning mechanism and a warning-control mapping matrix to achieve risk hierarchical response and intelligent collaborative control.

[0011] The digital twin interactive verification module builds a high-precision digital twin model of the rotating bridge, and realizes virtual verification and dynamic simulation of the control strategy through virtual and real two-way data interaction and real-time comparative analysis.

[0012] Furthermore, the multi-source sensor integration module includes the following components.

[0013] Multi-type sensor units integrate GNSS / RTK, three-axis accelerometer, inclinometer, displacement sensor, fiber Bragg grating and strain gauge to achieve comprehensive monitoring of bridge displacement, angle, acceleration and strain.

[0014] High-precision time synchronization unit, using PTP / IEEE 1588 precise time protocol and distributed clock synchronization technology, ensures accurate and unified timestamps of sensor data at different locations.

[0015] The energy management unit, combined with intelligent power management, ensures long-term stable operation of the sensor network in harsh environments.

[0016] The network communication and redundant backup unit adopts a hybrid wired and wireless communication method to achieve industrial-grade network transmission reliability, and ensures the continuity of data transmission through redundant design of communication paths.

[0017] The sensor hardware self-diagnosis unit realizes real-time monitoring of the sensor hardware status and periodic self-calibration, and automatically identifies hardware-level physical faults to ensure the reliability of the collected data source.

[0018] The environmental adaptability protection unit provides waterproof, dustproof, shockproof, anti-electromagnetic interference and temperature compensation protection measures for various sensors to ensure stable operation in harsh environments.

[0019] The sensor layout optimization unit determines the optimal sensor layout plan for key parts of the bridge, so as to obtain the most comprehensive structural information with the least number of sensors.

[0020] Furthermore, the GNSS / RTK positioning accuracy in the multi-type sensor units is better than 1 cm, the three-axis accelerometer range is ±16g and the sensitivity is not less than 0.001g, the inclination sensor measurement range is ±90° and the accuracy is better than 0.01°, the displacement sensor measurement range is 0-500 mm and the accuracy is better than 0.1 mm, and the fiber grating strain sensor measurement accuracy is better than 1 microstrain.

[0021] Furthermore, the edge computing and data fusion module includes the following components.

[0022] Distributed edge computing units are deployed in key monitoring areas of bridges to enable on-site processing of sensor data.

[0023] The data preprocessing and noise filtering unit, based on the multi-scale Kalman filtering algorithm, handles the noise problem in the sensor raw data and improves the quality of the single-source signal.

[0024] The data quality assessment unit automatically identifies and repairs data-level outliers, missing values, and mutation values ​​in sensor data through statistical learning and pattern recognition methods to ensure the reliability of data fusion.

[0025] The computing resource dynamic scheduling unit intelligently allocates and schedules edge computing resources according to the monitoring task priority and real-time data traffic to ensure the real-time and overall efficiency of key data processing.

[0026] The multi-source data fusion unit dynamically selects appropriate data fusion algorithms for different monitoring scenarios to achieve comprehensive analysis and information extraction of multi-source heterogeneous data.

[0027] The edge-cloud data distribution unit establishes a layered computing architecture and data transmission mechanism between edge nodes and cloud platforms. Lightweight real-time tasks are completed at the edge, and complex analysis tasks are pushed to the cloud.

[0028] The data compression and transmission optimization unit adopts a context-aware compression algorithm to dynamically adjust the compression ratio according to data importance and bandwidth conditions, thereby optimizing data transmission efficiency while ensuring monitoring accuracy.

[0029] Furthermore, the multi-physics coupling modeling module includes the following components.

[0030] The structural mechanics modeling unit establishes a refined finite element model of the rotating bridge and calculates the stress distribution and structural deformation under various loads.

[0031] The temperature field coupling unit simulates the non-uniform thermal expansion effects of sunlight radiation and ambient temperature changes on various bridge components, and calculates the influence of thermal stress and thermal deformation on the bridge posture.

[0032] The contact friction mechanics unit accurately models the friction characteristics and wear state of key contact interfaces and calculates the damping effect of friction torque on the rotation process.

[0033] The fluid-structure interaction unit simulates the dynamic effects of wind loads and water flow on the bridge and predicts attitude fluctuations by calculating fluid-induced vibrations and aerodynamic effects.

[0034] The physical model boundary condition unit continuously calibrates the model's physical boundary constraints and load conditions based on sensor monitoring data to ensure the physical consistency of the model.

[0035] The multi-scale computing collaborative unit integrates the multi-scale analysis technology of macroscopic structural response and microscopic material behavior to improve the physical accuracy of the model while ensuring computational efficiency.

[0036] Furthermore, the temperature field coupling unit is configured to achieve: ambient temperature monitoring range -40℃ to 85℃, with an accuracy better than ±0.5℃; solar radiation intensity monitoring range 0-1200W / m², with an accuracy better than ±5%; support 24-hour temperature field evolution prediction, with a prediction error less than ±2.5℃; thermal deformation calculation accuracy better than ±5% of the measured value; automatic identification of high-risk areas with temperature gradients exceeding 15℃ / m, and issuance of early warnings.

[0037] Furthermore, the adaptive control strategy module includes the following components.

[0038] The predictive control unit predicts the future state response of the bridge based on the multi-physics field coupling model, calculates the optimal control instruction sequence through rolling time domain optimization, and realizes feedforward control of the rotation process.

[0039] The control parameter online identification unit continuously updates the control-related dynamic parameters using real-time monitoring data, capturing changes in structural characteristics, friction coefficients, and external loads.

[0040] The reinforcement learning optimization unit extracts the optimal strategy from historical control experience, adapts to complex environmental changes and model uncertainties, and continuously improves the control effect.

[0041] The control scheme optimization unit balances the rotation accuracy, energy consumption, execution time and safety margin objectives, and generates a control scheme with the best overall performance through the Pareto optimal method.

[0042] The robustness assurance unit designs control laws that are resistant to parameter disturbances and external interference, ensuring control stability in harsh environments and when some sensors fail.

[0043] The actuator distribution unit optimizes the distribution of control forces based on the characteristics and status of the bridge's multiple drivers to avoid actuator overload and structural stress concentration.

[0044] The control strategy performance evaluation unit conducts real-time evaluation and quantitative analysis of the implementation effect of the control strategy, providing a basis for the iterative optimization of the control strategy.

[0045] Furthermore, the hierarchical warning and collaborative control module includes the following components.

[0046] The multi-dimensional condition assessment unit comprehensively analyzes structural stress, displacement, vibration and environmental condition parameters, and calculates bridge health status indicators in real time.

[0047] The warning level division unit establishes a five-level warning threshold system of normal, attention, warning, danger and emergency based on bridge safety standards and historical data, so as to achieve accurate classification of abnormal conditions.

[0048] The early warning-control mapping matrix unit establishes a dynamic mapping relationship between early warning levels and control strategies, and automatically matches corresponding control parameters and execution plans for different levels of risks.

[0049] The multi-system collaborative unit coordinates the linkage response of the swivel mechanism, shock absorption system and locking device to ensure that the overall system can coordinately respond to abnormal conditions of different levels.

[0050] The human-machine interactive confirmation unit provides manual intervention and decision confirmation mechanisms for high-level warning events, combining operator experience with automatic control to form dual protection.

[0051] The emergency downgrade control unit automatically starts the downgraded operation mode in the event of partial failure of the sensor or controller, maintaining basic control functions and ensuring the safety of the swing bridge.

[0052] The early warning response recording unit records the triggering conditions of the early warning events, the system response process and the event processing results, providing data support for the subsequent improvement of the early warning mechanism.

[0053] Furthermore, the digital twin interaction verification module includes the following components.

[0054] The high-precision geometric modeling unit constructs a digital twin model of the rotating bridge based on laser scanning and BIM technology, achieving accurate mapping of the physical structure to the virtual model.

[0055] The integrated physical environment simulation unit builds a comprehensive simulation environment including mechanics, thermals, and fluid based on multi-physics field models, which serves as the basic platform for virtual testing and verification.

[0056] The physical-virtual synchronization unit realizes real-time mapping of monitoring data to the virtual model through a dedicated data interface, maintaining the dynamic consistency between the digital twin and the physical bridge status.

[0057] The control strategy virtual verification unit uses the digital twin environment to pre-test the control strategy, evaluate its safety and effectiveness, and provide suggestions for the optimization of the actual control solution.

[0058] The virtual-reality difference analysis unit compares the deviation between the digital twin prediction results and the actual monitoring data, identifies model defects and potential anomalies, and continuously optimizes the accuracy of the digital twin model.

[0059] The multi-scenario rehearsal unit simulates the rotation behavior of the bridge under special working conditions, formulates emergency plans in advance, and enhances the ability to deal with complex situations.

[0060] The visual interaction unit intuitively displays the bridge status and control process through a three-dimensional visual interface and augmented reality technology, supporting engineers in immersive analysis and decision-making assistance.

[0061] Furthermore, the digital twin model constructed by the high-precision geometric modeling unit has the following characteristics.

[0062] The geometric accuracy of key parts is better than ±5mm, and the overall geometric accuracy is better than ±20mm.

[0063] It contains no less than 300 independent components and 60 key connection nodes.

[0064] The total model data volume does not exceed 5GB, ensuring smooth operation on ordinary workstations.

[0065] Supports at least 3 standard data exchange formats, including IFC, STEP and STL.

[0066] It adopts parametric modeling technology to support rapid model updates, and the response time for single parameter modification is less than 10 seconds.

[0067] The data synchronization delay with the BIM system is less than 5 minutes.

[0068] Supports dynamic switching of three levels of detail (LOD200-LOD400) to balance display accuracy and operating efficiency.

[0069] Compared with the prior art, the present application has the following beneficial effects.

[0070] This application is based on multi-source sensor fusion, edge computing, multi-physics field coupling modeling and digital twin technology to construct a high-precision, real-time adaptive rotating bridge posture monitoring and control system to achieve efficient, accurate and safe dynamic adjustment and intelligent collaborative control of the bridge posture. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is a structural schematic diagram of a rotating bridge posture monitoring system based on dynamic feedback control disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0072] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described embodiments are part of the embodiments of the present invention, not all of the embodiments.

[0073] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0074] The embodiments and directional terms described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0075] like Figure 1 As shown, a rotating bridge posture monitoring system based on dynamic feedback control includes the following modules.

[0076] The multi-source sensor integration module integrates multiple types of sensor devices, builds an all-round posture monitoring network, and realizes high-precision real-time monitoring of key parts of the bridge.

[0077] The edge computing and data fusion module uses the edge computing nodes deployed on-site and the multi-scale Kalman filtering algorithm to realize real-time on-site processing and fusion of sensor data, effectively reducing data transmission delays and improving monitoring stability.

[0078] The multi-physics coupling modeling module builds a multi-physics coupling model including structural mechanics, temperature field, contact friction and fluid action to achieve accurate prediction of the posture changes of the rotating bridge.

[0079] The adaptive control strategy module dynamically optimizes control parameters based on real-time monitoring data, combining model predictive control, parameter adaptation and reinforcement learning technology to achieve high-precision adaptive control of the rotating bridge posture.

[0080] The hierarchical warning and collaborative control module automatically matches the corresponding control strategies by establishing a multi-level warning mechanism and a warning-control mapping matrix to achieve risk hierarchical response and intelligent collaborative control.

[0081] The digital twin interactive verification module builds a high-precision digital twin model of the rotating bridge, and realizes virtual verification and dynamic simulation of the control strategy through virtual and real two-way data interaction and real-time comparative analysis.

[0082] In this embodiment, the multi-source sensor integration module serves as the basic data acquisition part of the entire rotating bridge posture monitoring system, and its role is mainly reflected in the comprehensiveness, accuracy and real-time monitoring. By integrating multiple types of sensing equipment, the key parts of the bridge can be monitored from different angles and dimensions, and a comprehensive posture monitoring network can be constructed. For example, the strain gauges installed at key parts such as the piers, abutments, and beams of the bridge can sense the strain of the structure in real time during the rotation process, and then deduce the stress state of the structure; the accelerometer can capture the vibration acceleration of the bridge during dynamic rotation, and provide data support for subsequent structural dynamic analysis; the inclination sensor directly obtains the inclination angle information of the bridge components, and the displacement meter can accurately measure the displacement changes during the rotation process. This combination of multiple types of sensors enables the monitoring data to complement and verify each other, greatly improving the comprehensiveness and accuracy of bridge posture monitoring. Moreover, these sensors can collect data in real time at a high frequency, ensuring that the system can capture the slight changes in the posture of the rotating bridge in a timely manner, and provide a timely and accurate data basis for subsequent dynamic feedback control, thereby effectively ensuring the safety and stability of the rotating bridge during construction and operation.

[0083] In this embodiment, the edge computing and data fusion module plays a key role in the swivel bridge posture monitoring system, and its technical effect is remarkable. On the one hand, the edge computing nodes deployed on site can realize real-time on-site processing of sensor data. Compared with the traditional method of transmitting all sensor data to the cloud or central server for processing, edge computing greatly reduces the delay of data transmission. This is because the edge computing node is close to the data source, and can perform preliminary processing on a large amount of collected sensor data, such as data cleaning, filtering, feature extraction, etc., eliminate invalid or redundant data, and only transmit valuable key data to the subsequent processing module, thereby effectively reducing the burden of data transmission and improving the response speed of the entire system. On the other hand, the application of the multi-scale Kalman filter algorithm further improves the effect of data fusion. The algorithm can comprehensively consider the characteristics, accuracy and time scale differences of different sensor data, and perform fusion processing on multi-source sensor data. For example, for sensor data with different sampling frequencies, Kalman filtering can unify them to a suitable time scale for fusion, and at the same time, according to the reliability and relevance of each sensor, assign different weights to obtain a more accurate and stable state estimate. This not only improves the accuracy and reliability of monitoring data, but also enhances the system's anti-interference ability and stability in complex environments, provides high-quality data support for subsequent rotating bridge posture prediction and control, and ensures the efficient and stable operation of the entire monitoring system.

[0084] In this embodiment, the multi-physics field coupling modeling module is the core of realizing accurate prediction of the posture of the rotating bridge. Its role is reflected in the ability to comprehensively consider the influence of various physical factors on the bridge posture, thereby improving the accuracy and reliability of the prediction. In the actual working process, the rotating bridge is subject to the interaction of multiple physical fields. For example, the structural mechanics field determines the structural deformation and internal force distribution of the bridge under the action of its own gravity, vehicle load and inertia force during the rotation process; the influence of the temperature field is because the bridge structure will expand and contract under different temperature conditions, resulting in changes in size and shape, which in turn affects the posture; the contact friction field involves the change of friction in the rotating device, which will directly affect the smoothness and accuracy of the rotation; the fluid action field mainly considers the force of fluids such as wind and water flow on the bridge, especially in windy weather or near water environments, which will have a significant impact on the posture of the bridge. The multi-physics field coupling modeling module can simulate the posture changes of the rotating bridge under various complex working conditions by establishing a mathematical model containing these physical fields and using numerical methods such as finite element analysis to solve them. For example, when predicting the rotational posture of a bridge under different temperature gradients, the model can comprehensively consider the thermal expansion coefficient, temperature distribution law and structural mechanical properties of the structural material, and accurately calculate the deformation and posture change trend of each component. This precise prediction capability provides a scientific basis for formulating reasonable control strategies in advance, helps to achieve refined management of the posture of the rotating bridge, effectively prevent and avoid potential risks such as excessive posture deviation, and ensure the safety and accuracy of the rotation construction and operation.

[0085] In this embodiment, the adaptive control strategy module is the key link for the entire rotating bridge posture monitoring system to achieve dynamic feedback control. Its role is mainly reflected in the ability to dynamically optimize control parameters according to real-time monitoring data to achieve high-precision adaptive control. Based on the real-time acquired bridge posture monitoring data, the module combines model predictive control, parameter adaptation and reinforcement learning technology to automatically adjust the control strategy to adapt to the changes in the bridge posture. For example, the model predictive control technology can predict the posture change trend in the future period of time according to the current posture state and the multi-physics field coupling model, and then plan the optimal control input in advance, such as adjusting the driving force and braking force of the rotating device, so that the bridge posture changes according to the expected trajectory; the parameter adaptive technology can identify the changes in system parameters in real time, such as structural stiffness, friction coefficient, etc. When these parameters change due to environmental changes, structural damage and other factors, the corresponding parameters in the control algorithm are automatically adjusted to ensure that the control performance is not affected; reinforcement learning technology enables the control system to autonomously optimize the control strategy in a complex environment through continuous trial and error learning to achieve the best posture control effect. This adaptive control strategy can effectively cope with various uncertainties faced by the rotating bridge during its actual operation, such as sudden changes in wind speed and direction, drastic temperature fluctuations, dynamic changes in vehicle loads, etc., ensuring that the rotating bridge always operates in a safe, stable and precise posture state. It greatly improves the intelligence level and control accuracy of the entire system, which is of great significance for ensuring the construction quality and operational safety of the rotating bridge.

[0086] In this embodiment, the hierarchical warning and collaborative control module plays an important role in risk management and multi-component collaborative control in the rotating bridge posture monitoring system. Its role is reflected in the ability to achieve hierarchical response to risks and intelligent collaborative control. By establishing a multi-level warning mechanism, the module can divide the warning into different levels according to the abnormality of the posture monitoring data and the size of the potential risk. For example, the first-level warning indicates that the posture deviation is within the allowable range and the system is operating normally; the second-level warning indicates that the posture deviation is close to the safety threshold and needs to be paid attention to and adjusted locally; the third-level warning indicates that the posture has serious deviations, there are safety hazards, and emergency control measures need to be taken immediately. For different levels of warnings, the system automatically matches the corresponding control strategy through the warning-control mapping matrix to achieve intelligent collaborative control of the various components of the rotating bridge. For example, when a second-level warning occurs, the system may automatically adjust the driving power of the rotating device, change the position of the counterweight block, etc. to reduce the posture deviation; when a third-level warning occurs, the system will quickly start the braking device, stop the rotation action, and issue an alarm to notify relevant personnel to check and handle it. This hierarchical warning and collaborative control mechanism can effectively control risks in the bud and avoid further expansion of posture deviations that may cause safety accidents. At the same time, through the coordinated actions of various components, it improves the safety and reliability of the entire rotating bridge system, ensures its stable operation under complex working conditions, and provides solid safety guarantees for the rotation construction and operation.

[0087] In this embodiment, the digital twin interactive verification module provides a powerful tool for the optimization and virtual verification of the control strategy of the rotating bridge posture monitoring system. Its role is mainly reflected in the virtual verification and dynamic simulation of the control strategy through virtual and real two-way data interaction and real-time comparative analysis. The module first constructs a high-precision digital twin model of the rotating bridge. This model is not only consistent with the actual bridge in geometry, but also integrates its physical properties, mechanical properties, control logic and other information, which can truly reflect the operating status and behavior characteristics of the actual bridge. During operation, the posture monitoring data of the actual bridge is transmitted to the digital twin model in real time, driving the model to synchronize updates and simulation operations. At the same time, the simulation results of the digital twin model are also fed back to the actual control system, forming a virtual and real two-way data interaction closed loop. Through this interaction, on the one hand, new control strategies can be pre-verified in a virtual environment, such as simulating different control parameter adjustment schemes on the digital twin model to observe their effects on posture control, avoiding the risks that may be caused by direct testing on the actual bridge; on the other hand, real-time comparison and analysis of the posture data of the actual bridge and the digital twin model can timely discover the deviation between the model and the actual situation, and then correct and optimize the model, so that the digital twin model always maintains a high-precision state, better serving the formulation and optimization of the control strategy. This interactive verification mechanism based on digital twins not only improves the scientificity and reliability of the control strategy, but also provides a safe and efficient test platform for the posture control of the rotating bridge, which helps to further improve the intelligence level and control accuracy of the entire monitoring system and promote the development and innovation of rotating bridge engineering technology.

[0088] In summary, the dynamic feedback control-based rotating bridge posture monitoring system realizes all-round, high-precision, and intelligent monitoring and control of the rotating bridge posture through the synergy of multiple modules such as multi-source sensor integration, edge computing and data fusion, multi-physics field coupling modeling, adaptive control strategy, hierarchical warning and collaborative control, and digital twin interactive verification. From the comprehensiveness and accuracy of basic data acquisition, to the efficiency and stability of data processing, to the scientificity and reliability of posture prediction, as well as the adaptive optimization of control strategies, hierarchical response and collaborative regulation of risks, and finally to the virtual verification and dynamic simulation of control strategies, each module plays an indispensable and key role, and they cooperate and complement each other to form a complete, efficient, and intelligent rotating bridge posture monitoring system.

[0089] Furthermore, the multi-source sensor integration module includes the following components.

[0090] Multi-type sensor units integrate GNSS / RTK, three-axis accelerometer, inclinometer, displacement sensor, fiber Bragg grating and strain gauge to achieve comprehensive monitoring of bridge displacement, angle, acceleration and strain.

[0091] High-precision time synchronization unit, using PTP / IEEE 1588 precise time protocol and distributed clock synchronization technology, ensures accurate and unified timestamps of sensor data at different locations.

[0092] The energy management unit, combined with intelligent power management, ensures long-term stable operation of the sensor network in harsh environments.

[0093] The network communication and redundant backup unit adopts a hybrid wired and wireless communication method to achieve industrial-grade network transmission reliability, and ensures the continuity of data transmission through redundant design of communication paths.

[0094] The sensor hardware self-diagnosis unit realizes real-time monitoring of the sensor hardware status and periodic self-calibration, and automatically identifies hardware-level physical faults to ensure the reliability of the collected data source.

[0095] The environmental adaptability protection unit provides waterproof, dustproof, shockproof, anti-electromagnetic interference and temperature compensation protection measures for various sensors to ensure stable operation in harsh environments.

[0096] The sensor layout optimization unit determines the optimal sensor layout plan for key parts of the bridge, so as to obtain the most comprehensive structural information with the least number of sensors.

[0097] In summary, the multi-source sensor integration module integrates multiple types of sensor units, including GNSS / RTK, three-axis accelerometers, inclination sensors, displacement sensors, fiber Bragg gratings and strain gauges, to achieve comprehensive monitoring of bridge displacement, angle, acceleration and strain, ensuring the comprehensiveness of data collection. The high-precision time synchronization unit adopts PTP / IEEE 1588 precise time protocol and distributed clock synchronization technology to accurately unify the data timestamps of sensors at different locations, providing an accurate time reference for subsequent data fusion and analysis. The energy management unit combines intelligent power management technology to ensure the long-term stable operation of the sensor network in harsh environments and enhance the reliability and adaptability of the system. The network communication and redundant backup unit adopts a hybrid communication mode of wired and wireless, and through the redundant design of the communication path, it ensures the high reliability and continuity of data transmission and avoids data loss or interruption. The sensor hardware self-diagnosis unit can monitor the sensor hardware status in real time, realize periodic self-calibration, automatically identify hardware-level physical faults, and ensure the reliability of collected data from the source. The environmental adaptability protection unit provides various sensors with multiple protection measures such as waterproof, dustproof, shockproof, anti-electromagnetic interference and temperature compensation, ensuring that the sensors can still work stably in harsh environments and expanding the scope of application of the system. The sensor layout optimization unit determines the optimal sensor layout plan for key parts of the bridge through scientific layout planning, and achieves the most comprehensive structural information with the least sensors, improving the economy and efficiency of the system. The synergy of these components enables the multi-source sensor integration module to efficiently, stably and accurately provide high-quality monitoring data for the rotating bridge posture monitoring system, providing a solid foundation for subsequent analysis, prediction and control.

[0098] Furthermore, the GNSS / RTK positioning accuracy in the multi-type sensor units is better than 1 cm, the three-axis accelerometer range is ±16g and the sensitivity is not less than 0.001g, the inclination sensor measurement range is ±90° and the accuracy is better than 0.01°, the displacement sensor measurement range is 0-500 mm and the accuracy is better than 0.1 mm, and the fiber grating strain sensor measurement accuracy is better than 1 microstrain.

[0099] In summary, the GNSS / RTK positioning accuracy of the multi-type sensor units is better than 1 cm, which can provide high-precision displacement monitoring for the rotating bridge and ensure the precise control of the bridge position during the rotation process. The three-axis accelerometer has a range of ±16g and a sensitivity of not less than 0.001g. It can accurately measure the acceleration changes of the bridge in three-dimensional space. It has high sensitivity and can effectively capture the tiny vibrations and dynamic responses of the bridge, providing key data for structural health monitoring. The inclination sensor has a measurement range of ±90° and an accuracy of better than 0.01°. It can accurately measure the angle changes of bridge components and ensure the precise adjustment of the posture during the rotation process. The displacement sensor has a measurement range of 0-500 mm and an accuracy of better than 0.1 mm. It can directly measure the displacement changes of key parts of the rotating bridge and provide accurate displacement feedback for posture control. The fiber Bragg grating strain sensor has a measurement accuracy of better than 1 microstrain, which can accurately monitor the strain of the bridge structure, detect structural stress anomalies in a timely manner, and ensure the safety of the bridge. The synergistic effect of these high-precision sensors provides all-round, high-precision and high-reliability data support for the rotating bridge posture monitoring system, ensuring the safety and stability of the rotating bridge during construction and operation.

[0100] Furthermore, the optimal sensor layout scheme for the key parts of the bridge is determined to obtain the most comprehensive structural information with the least number of sensors, including the following steps.

[0101] Through sensitivity analysis, the response change rate of stress and displacement at each position to external influences is calculated, sensitivity indicators are defined, and the areas with the greatest impact on the change of bridge posture are screened.

[0102] Select high-sensitivity areas as candidate monitoring points, based on the optimization objective function min x ∑ i=1 n w i ·S i (x) Determine the preliminary sensor layout and set the sensor quantity constraint, where x is the decision variable for sensor layout; w i is the weight coefficient of sensor position i; S i (x) is the sensitivity function at position i, which indicates the sensitivity of this position to the change of bridge response after the sensor is deployed.

[0103] Calculate information entropy H(X)=-∑ i=1 n p(x i )logp(x i ) evaluates the layout information richness and optimizes the sensor position by maximizing the information entropy, where H(X) is the information entropy, which is used to measure the information richness after the sensor is arranged; p(x i ) is the sensor at position x iThe information probability distribution represents the distribution of the response information at the sensor position i.

[0104] Genetic algorithms, simulated annealing or integer programming methods are used to solve the optimization problem, complete the final sensor layout, and verify and adjust the solution based on finite element simulation.

[0105] Continuously update sensitivity analysis based on actual monitoring data, adjust sensor positions or increase points in key areas to ensure the stability and accuracy of long-term monitoring.

[0106] In summary, the sensor layout optimization unit calculates the response change rate of stress and displacement at each position to external influences through sensitivity analysis, defines sensitivity indicators, screens the areas with the greatest impact on bridge posture changes, selects high-sensitivity areas as candidate monitoring points, and determines the preliminary sensor layout based on the optimization objective function, while setting the sensor quantity constraint. Then, the information entropy is calculated to evaluate the information richness of the layout, the sensor position is optimized by maximizing the information entropy, and the optimization problem is solved by genetic algorithms, simulated annealing or integer programming methods to complete the final sensor layout, and the scheme is verified and adjusted based on finite element simulation. Finally, the sensitivity analysis is continuously updated according to the actual monitoring data, and the sensor position is adjusted or the key area points are increased to ensure the stability and accuracy of long-term monitoring.

[0107] Furthermore, the edge computing and data fusion module includes the following components.

[0108] Distributed edge computing units are deployed in key monitoring areas of bridges to enable on-site processing of sensor data.

[0109] The data preprocessing and noise filtering unit, based on the multi-scale Kalman filtering algorithm, handles the noise problem in the sensor raw data and improves the quality of the single-source signal.

[0110] The data quality assessment unit automatically identifies and repairs data-level outliers, missing values, and mutation values ​​in sensor data through statistical learning and pattern recognition methods to ensure the reliability of data fusion.

[0111] The computing resource dynamic scheduling unit intelligently allocates and schedules edge computing resources according to the monitoring task priority and real-time data traffic to ensure the real-time and overall efficiency of key data processing.

[0112] The multi-source data fusion unit dynamically selects appropriate data fusion algorithms for different monitoring scenarios to achieve comprehensive analysis and information extraction of multi-source heterogeneous data.

[0113] The edge-cloud data distribution unit establishes a layered computing architecture and data transmission mechanism between edge nodes and cloud platforms. Lightweight real-time tasks are completed at the edge, and complex analysis tasks are pushed to the cloud.

[0114] The data compression and transmission optimization unit adopts a context-aware compression algorithm to dynamically adjust the compression ratio according to data importance and bandwidth conditions, thereby optimizing data transmission efficiency while ensuring monitoring accuracy.

[0115] In summary, the edge computing and data fusion module realizes on-site processing of sensor data and reduces the amount of data transmission through distributed edge computing units. The data preprocessing and noise filtering unit adopts a multi-scale Kalman filtering algorithm to effectively remove noise from the raw data of the sensor and improve the signal quality. The data quality assessment unit uses statistical learning and pattern recognition methods to automatically identify and repair abnormal, missing and mutant values ​​in the data to ensure the reliability of the data. The computing resource dynamic scheduling unit intelligently allocates edge computing resources according to the monitoring task priority and real-time data traffic to ensure real-time processing of key data. The multi-source data fusion unit dynamically selects fusion algorithms according to the monitoring scenario to achieve comprehensive analysis of multi-source heterogeneous data. The edge cloud data distribution unit establishes a hierarchical computing architecture, lightweight tasks are completed at the edge, and complex tasks are pushed to the cloud. The data compression and transmission optimization unit adopts a context-aware compression algorithm to dynamically adjust the compression ratio according to the importance of the data and the bandwidth status, and optimizes the transmission efficiency while ensuring accuracy. The synergy of these components improves the efficiency and quality of data processing, reduces data transmission delay and cost, enhances the stability and adaptability of the system, and provides efficient data support for the rotating bridge posture monitoring system.

[0116] Furthermore, the multi-physics coupling modeling module includes the following components.

[0117] The structural mechanics modeling unit establishes a refined finite element model of the rotating bridge and calculates the stress distribution and structural deformation under various loads.

[0118] The temperature field coupling unit simulates the non-uniform thermal expansion effects of sunlight radiation and ambient temperature changes on various bridge components, and calculates the influence of thermal stress and thermal deformation on the bridge posture.

[0119] The contact friction mechanics unit accurately models the friction characteristics and wear state of key contact interfaces and calculates the damping effect of friction torque on the rotation process.

[0120] The fluid-structure interaction unit simulates the dynamic effects of wind loads and water flow on the bridge and predicts attitude fluctuations by calculating fluid-induced vibrations and aerodynamic effects.

[0121] The physical model boundary condition unit continuously calibrates the model's physical boundary constraints and load conditions based on sensor monitoring data to ensure the physical consistency of the model.

[0122] The multi-scale computing collaborative unit integrates the multi-scale analysis technology of macroscopic structural response and microscopic material behavior to improve the physical accuracy of the model while ensuring computational efficiency.

[0123] In summary, the multi-physics coupling modeling module establishes a refined finite element model of the rotating bridge through the structural mechanics modeling unit, and calculates the stress distribution and structural deformation under various loads. The temperature field coupling unit simulates the non-uniform thermal expansion effect of sunlight radiation and ambient temperature changes on bridge components, and calculates the influence of thermal stress and thermal deformation on the posture. The contact friction mechanics unit accurately models the friction characteristics and wear state of key contact interfaces, and calculates the damping effect of friction torque on the rotation process. The fluid-structure interaction unit simulates the dynamic effects of wind loads and water flow on the bridge and predicts posture fluctuations. The physical model boundary condition unit continuously calibrates the physical boundary constraints and load conditions of the model according to sensor data to ensure the physical consistency of the model. The multi-scale computing synergy unit integrates the multi-scale analysis technology of macroscopic structural response and microscopic material behavior to improve the physical accuracy of the model while ensuring computational efficiency. The synergy of these components enables the multi-physics coupling modeling module to comprehensively consider the influence of various physical factors on the posture of the rotating bridge, realize accurate prediction of the posture change of the rotating bridge, provide a scientific basis for subsequent control strategies, and ensure the construction quality and operation safety of the rotating bridge.

[0124] Furthermore, the temperature field coupling unit is configured to achieve: ambient temperature monitoring range -40℃ to 85℃, with an accuracy better than ±0.5℃; solar radiation intensity monitoring range 0-1200W / m², with an accuracy better than ±5%; support 24-hour temperature field evolution prediction, with a prediction error less than ±2.5℃; thermal deformation calculation accuracy better than ±5% of the measured value; automatic identification of high-risk areas with temperature gradients exceeding 15℃ / m, and issuance of early warnings.

[0125] In summary, the temperature field coupling unit can accurately calculate thermal deformation, automatically identify high-risk areas and issue warnings through high-precision monitoring of ambient temperature and solar radiation intensity, combined with 24-hour temperature field evolution prediction, providing accurate temperature field data support for the rotating bridge posture monitoring system, ensuring the safety and stability of the system under different temperature conditions. Specifically, its ambient temperature monitoring range is -40℃ to 85℃, with an accuracy better than ±0.5℃; the solar radiation intensity monitoring range is 0-1200W / m², with an accuracy better than ±5%. The unit can support 24-hour temperature field evolution prediction, with a prediction error of less than ±2.5℃, and thermal deformation calculation accuracy better than ±5% of the measured value. It can also automatically identify high-risk areas with temperature gradients exceeding 15℃ / m, and issue warnings in a timely manner, thereby effectively preventing abnormal posture of the rotating bridge caused by temperature changes.

[0126] Furthermore, a refined finite element model of the rotating bridge is established to calculate the stress distribution and structural deformation under various loads, including the following steps.

[0127] The rotating bridge structure is discretized into finite element units, the node and unit geometric parameters are defined, a refined finite element mesh is established, and material properties and section information are input.

[0128] The local stiffness matrix of each unit is calculated using shape functions and strain-displacement matrix [B], and globally assembled to form the overall stiffness matrix [K]. At the same time, the external force vector [F] including dead load, live load, temperature stress and wind load is constructed.

[0129] The node displacement vector is obtained by solving the finite element equilibrium equation [K]·[u]=[F], and then the strain and stress distribution of each unit is calculated by ε=[B]·[u] and σ=[D]·ε, where [u] is the node displacement vector, which represents the translation and rotation displacement of each node of the bridge; ε is the strain tensor, which represents the degree of deformation of the material under the action of external force; σ is the stress tensor, which represents the internal force distribution caused by external force inside the material; and [D] is the elastic matrix of the material.

[0130] Using the thermal expansion formula ΔL=αL 0 ΔT calculates the local deformation caused by temperature and uses the friction torque formula M friction =μ·N·r Considering the contact friction effect during the rotation process, the deformation result is updated, where ΔL is the change in structure length caused by temperature change; α is the linear expansion coefficient of the material, which indicates the degree of expansion of the material when the temperature changes; L 0 is the initial length of the structure; ΔT is the temperature change, indicating the change of ambient temperature; M friction is the friction torque, which indicates the torque generated by the friction of the contact surface; μ is the friction coefficient, which indicates the friction resistance between the contact surfaces; N is the normal contact force, which indicates the normal pressure on the contact surface; r is the length of the lever arm, which indicates the distance from the force application point to the rotation axis.

[0131] The fluid force F generated by wind load or water flow fluid =C d ·(1 / 2)ρv 2 A is included in the calculation to realize the superposition effect of various loads and calculate the overall deformation Δx total =Δx load +Δx thermal +Δx friction +Δx fluid , where F fluid is the fluid force, which represents the force of wind load or water flow on the bridge; C dis the drag coefficient, which indicates the magnitude of the resistance encountered by an object in a fluid; ρ is the density of the fluid, which indicates the ratio of the mass to the volume of the fluid; v is the velocity of the fluid, which indicates the velocity of the fluid flow; A indicates the intersection area between the fluid action surface and the fluid flow direction; Δx total is the total deformation of the bridge, including the deformation under all loads; Δx load is the deformation caused by external load; Δx thermal is the deformation caused by temperature effect; Δx friction is the deformation caused by friction torque; Δx fluid is the deformation caused by wind load.

[0132] In summary, a refined finite element model of the rotating bridge is established. By discretizing the structure, defining geometric parameters and inputting material properties, the stiffness matrix and external force vector are calculated using shape functions and strain-displacement matrices, and the equilibrium equations are solved to obtain the node displacements and stress-strain distributions. At the same time, the coupling effects of multiple physical fields such as temperature, friction torque and fluid are considered to achieve accurate calculation of the stress distribution and structural deformation of the rotating bridge under various loads, providing a scientific basis for the safety assessment and design optimization of the bridge.

[0133] Furthermore, the adaptive control strategy module includes the following components.

[0134] The predictive control unit predicts the future state response of the bridge based on the multi-physics field coupling model, calculates the optimal control instruction sequence through rolling time domain optimization, and realizes feedforward control of the rotation process.

[0135] The control parameter online identification unit continuously updates the control-related dynamic parameters using real-time monitoring data, capturing changes in structural characteristics, friction coefficients, and external loads.

[0136] The reinforcement learning optimization unit extracts the optimal strategy from historical control experience, adapts to complex environmental changes and model uncertainties, and continuously improves the control effect.

[0137] The control scheme optimization unit balances the rotation accuracy, energy consumption, execution time and safety margin objectives, and generates a control scheme with the best overall performance through the Pareto optimal method.

[0138] The robustness assurance unit designs control laws that are resistant to parameter disturbances and external interference, ensuring control stability in harsh environments and when some sensors fail.

[0139] The actuator distribution unit optimizes the distribution of control forces based on the characteristics and status of the bridge's multiple drivers to avoid actuator overload and structural stress concentration.

[0140] The control strategy performance evaluation unit conducts real-time evaluation and quantitative analysis of the implementation effect of the control strategy, providing a basis for the iterative optimization of the control strategy.

[0141] In summary, the adaptive control strategy module predicts the future state of the bridge based on the multi-physics field coupling model through the prediction control unit to achieve feedforward control; the control parameter online identification unit updates the dynamic parameters in real time to adapt to the changes in structural characteristics and external loads; the reinforcement learning optimization unit extracts the optimal strategy from historical experience to deal with complex environments and model uncertainties; the control scheme optimization unit balances the rotation accuracy, energy consumption and other goals to generate the best comprehensive performance scheme; the robustness guarantee unit designs the anti-interference control law to ensure stable control in harsh environments; the actuator allocation unit optimizes the distribution of control force to avoid actuator overload and stress concentration; the control strategy performance evaluation unit evaluates the implementation effect in real time to provide a basis for iterative optimization. These components work together to achieve high-precision adaptive control of the rotation bridge posture and ensure its safe, stable and efficient operation.

[0142] Furthermore, the prediction control unit adopts a hierarchical predictive control algorithm with the following characteristics: the prediction time domain range is adjustable from 3 to 20 seconds; the control time domain resolution is better than 200ms; the state prediction error is less than ±3%; the calculation delay is less than 1000ms; and it supports no less than 2 rolling optimization algorithms, including quadratic programming and gradient descent.

[0143] In summary, the prediction control unit adopts a hierarchical prediction control algorithm, with an adjustable 3-20 second prediction time domain and a control time domain resolution better than 200ms, ensuring rapid response and accurate tracking of the rotating bridge posture changes. Its state prediction error is less than ±3%, and the calculation delay is less than 1000ms, ensuring the timeliness and accuracy of the control instructions. It supports no less than 2 rolling optimization algorithms, including quadratic programming and gradient descent method, and can flexibly select optimization strategies according to different working conditions and control objectives, to achieve efficient and accurate control of the rotating bridge posture, and improve the overall performance and adaptability of the system.

[0144] Furthermore, the reinforcement learning optimization unit adopts a deep deterministic policy gradient algorithm and a double Q-learning strategy, including.

[0145] The dimension of the state space is no less than 20.

[0146] The dimension of the action space is no less than 8 dimensions.

[0147] The neural network structure uses at least 3 layers of deep network, with no less than 64 neurons in each layer.

[0148] The reward function comprehensively considers three key indicators: position accuracy, energy consumption and safety.

[0149] The learning rate is dynamically adjusted in the range of 0.0005-0.005.

[0150] The experience replay cache capacity is no less than 50,000, and the sampling strategy combines priority and random sampling.

[0151] The number of training iterations is no less than 80,000, and the control performance is improved by at least 15% after convergence.

[0152] In summary, the reinforcement learning optimization unit adopts a deep deterministic policy gradient algorithm and a dual Q-learning strategy, and has high-dimensional processing capabilities with a state space dimension of no less than 20 dimensions and an action space dimension of no less than 8 dimensions. The neural network structure adopts at least 3 layers of deep networks and no less than 64 neurons in each layer to ensure the capture of complex relationships. The reward function comprehensively considers the three key indicators of position accuracy, energy consumption and safety. The dynamic adjustment range of the learning rate is 0.0005-0.005, the experience replay cache capacity is no less than 50,000, the sampling strategy combines priority and random sampling, the number of training iterations is no less than 80,000 times, and the control performance is improved by at least 15% after convergence. These features enable the reinforcement learning optimization unit to adapt to complex environmental changes and model uncertainties, extract the optimal strategy from historical control experience, and continuously improve the control effect, providing strong support for the high-precision adaptive control of the rotating bridge posture.

[0153] Furthermore, based on the characteristics and states of the multi-actuators of the bridge, the control force is optimally distributed, including the following steps.

[0154] Apply the formula Δθ(t)=θ target -θ(t) calculates the attitude error Δθ(t) as the control target, where θ target is the target posture of the bridge, that is, the desired posture state; θ(t) is the current actual posture of the bridge, which is the posture value obtained in real time during the actual monitoring process.

[0155] Introduce the contribution coefficient C of each driver to the posture change k , establish the model Δθ(t)=∑ k=1 K C k F k , used to describe the output control force F of each driver k Impact on bridge attitude adjustment, where K is the total number of actuators.

[0156] It is stipulated that the total control force must meet the overall demand of the bridge, that is, ∑ k=1 K F k =F total ; At the same time, each driver is limited by its maximum output F max,k , that is, 0≤F k ≤Fmax,k , where F total Represents the total control force required for the bridge.

[0157] Define the cost function J = γ∑ k=1 K F k 2 +β∥Δθ(t)∥ 2 , and by minimizing the cost function J, the optimal control force distribution is obtained to ensure that each actuator works together and prevents overload, where γ is the weight coefficient of the energy consumption term in the cost function; β is the weight coefficient of the attitude error term in the cost function.

[0158] In summary, efficient and accurate posture adjustment can be achieved by optimizing the distribution of control forces of multiple bridge drivers. First, the posture error is calculated based on the target posture and real-time monitoring posture, and the driver contribution coefficient is introduced to establish a mathematical model of control force and posture change. Secondly, ensure that the total control force meets the overall needs of the bridge, and constrain the output of a single driver. Finally, by minimizing the cost function that includes energy consumption and posture error, the optimal control force distribution is achieved, the driver collaborative work efficiency is improved, overload is prevented, and the bridge is ensured to be stably adjusted to the target posture.

[0159] Furthermore, the hierarchical warning and collaborative control module includes the following components.

[0160] The multi-dimensional condition assessment unit comprehensively analyzes structural stress, displacement, vibration and environmental condition parameters, and calculates bridge health status indicators in real time.

[0161] The warning level division unit establishes a five-level warning threshold system of normal, attention, warning, danger and emergency based on bridge safety standards and historical data, so as to achieve accurate classification of abnormal conditions.

[0162] The early warning-control mapping matrix unit establishes a dynamic mapping relationship between early warning levels and control strategies, and automatically matches corresponding control parameters and execution plans for different levels of risks.

[0163] The multi-system collaborative unit coordinates the linkage response of the swivel mechanism, shock absorption system and locking device to ensure that the overall system can coordinately respond to abnormal conditions of different levels.

[0164] The human-machine interactive confirmation unit provides manual intervention and decision confirmation mechanisms for high-level warning events, combining operator experience with automatic control to form dual protection.

[0165] The emergency downgrade control unit automatically starts the downgraded operation mode in the event of partial failure of the sensor or controller, maintaining basic control functions and ensuring the safety of the swing bridge.

[0166] The early warning response recording unit records the triggering conditions of the early warning events, the system response process and the event processing results, providing data support for the subsequent improvement of the early warning mechanism.

[0167] In summary, the hierarchical warning and collaborative control module analyzes the health status of the bridge in real time through the multi-dimensional status evaluation unit, and the warning level division unit establishes a five-level warning threshold system to accurately classify abnormal conditions. The warning-control mapping matrix unit dynamically matches the control strategy, and the multi-system collaborative unit links the rotation mechanism, shock absorption system and locking device to respond to abnormalities. The human-computer interaction confirmation unit provides manual intervention for high-level warnings, the emergency downgrade control unit maintains basic control functions in the event of local failure, and the warning response recording unit records the event processing process to optimize the mechanism. This module realizes real-time monitoring, risk classification, intelligent warning and collaborative control of the rotating bridge posture, ensures its safe operation, and improves the reliability and intelligence level of the system.

[0168] Furthermore, the multi-dimensional status assessment unit monitors in real time: displacement of no less than 12 key structural points; stress distribution of no less than 20 key cross sections; vibration acceleration of no less than 10 key positions; friction state of no less than 5 key contact surfaces; temperature field distribution of no less than 8 key positions. By comprehensively analyzing these multi-dimensional data, the bridge health status indicators are calculated in real time, providing comprehensive and accurate data support for the safety assessment and status monitoring of the bridge, ensuring the stability and safety of the rotating bridge under different working conditions.

[0169] Furthermore, the digital twin interaction verification module includes the following components.

[0170] The high-precision geometric modeling unit constructs a digital twin model of the rotating bridge based on laser scanning and BIM technology, achieving accurate mapping of the physical structure to the virtual model.

[0171] The integrated physical environment simulation unit builds a comprehensive simulation environment including mechanics, thermals, and fluid based on multi-physics field models, which serves as the basic platform for virtual testing and verification.

[0172] The physical-virtual synchronization unit realizes real-time mapping of monitoring data to the virtual model through a dedicated data interface, maintaining the dynamic consistency between the digital twin and the physical bridge status.

[0173] The control strategy virtual verification unit uses the digital twin environment to pre-test the control strategy, evaluate its safety and effectiveness, and provide suggestions for the optimization of the actual control solution.

[0174] The virtual-reality difference analysis unit compares the deviation between the digital twin prediction results and the actual monitoring data, identifies model defects and potential anomalies, and continuously optimizes the accuracy of the digital twin model.

[0175] The multi-scenario rehearsal unit simulates the rotation behavior of the bridge under special working conditions, formulates emergency plans in advance, and enhances the ability to deal with complex situations.

[0176] The visual interaction unit intuitively displays the bridge status and control process through a three-dimensional visual interface and augmented reality technology, supporting engineers in immersive analysis and decision-making assistance.

[0177] In summary, the digital twin interactive verification module builds an accurate digital twin model and integrated simulation environment through a high-precision geometric modeling unit and an integrated physical environment simulation unit, and uses a physical-virtual synchronization unit to achieve real-time data mapping to maintain dynamic consistency between virtual and real. The control strategy virtual verification unit pre-tests and evaluates the control strategy in the digital twin environment, the virtual-real difference analysis unit optimizes the model accuracy by comparing the predicted and actual data, the multi-scenario rehearsal unit simulates special working conditions and formulates emergency plans, and the visualization interaction unit intuitively displays the bridge status and control process through three-dimensional visualization and augmented reality technology. These components work together to achieve virtual verification and dynamic simulation of the swivel bridge posture monitoring system, improve the scientificity and reliability of the control strategy, enhance the ability to cope with complex situations, and promote the intelligent development of bridge engineering technology.

[0178] Furthermore, the digital twin model constructed by the high-precision geometric modeling unit has the following characteristics.

[0179] The geometric accuracy of key parts is better than ±5mm, and the overall geometric accuracy is better than ±20mm.

[0180] It contains no less than 300 independent components and 60 key connection nodes.

[0181] The total model data volume does not exceed 5GB, ensuring smooth operation on ordinary workstations.

[0182] Supports at least 3 standard data exchange formats, including IFC, STEP and STL.

[0183] It adopts parametric modeling technology to support rapid model updates, and the response time for single parameter modification is less than 10 seconds.

[0184] The data synchronization delay with the BIM system is less than 5 minutes.

[0185] Supports dynamic switching of three levels of detail (LOD200-LOD400) to balance display accuracy and operating efficiency.

[0186] In summary, the digital twin model constructed by the high-precision geometric modeling unit has high-precision characteristics, with geometric accuracy of key parts better than ±5mm and overall geometric accuracy better than ±20mm. It contains no less than 300 independent components and 60 key connection nodes, with rich and accurate details. The total data volume of the model does not exceed 5GB, ensuring smooth operation on ordinary workstations. It supports at least 3 standard data exchange formats (IFC, STEP and STL) and has strong compatibility. It adopts parametric modeling technology, supports fast updates, the response time for single parameter modification is less than 10 seconds, and the data synchronization delay with the BIM system is less than 5 minutes. It supports dynamic switching of three levels of detail from LOD200 to LOD400, balances display accuracy and operation efficiency, and provides efficient and accurate model support for virtual verification and dynamic simulation of the rotating bridge posture monitoring system.

[0187] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, a person skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A rotating bridge posture monitoring system based on dynamic feedback control, characterized in that: Includes the following modules: Multi-source sensor integration module integrates multiple types of sensor equipment to build a full-range posture monitoring network and achieve high-precision real-time monitoring of key parts of the bridge; The edge computing and data fusion module uses the edge computing nodes deployed on-site and the multi-scale Kalman filter algorithm to realize real-time on-site processing and fusion of sensor data, effectively reducing data transmission delays and improving monitoring stability; The multi-physics coupling modeling module builds a multi-physics coupling model including structural mechanics, temperature field, contact friction and fluid action to achieve accurate prediction of the posture change of the rotating bridge; The adaptive control strategy module dynamically optimizes control parameters based on real-time monitoring data, combining model predictive control, parameter adaptation and reinforcement learning technology to achieve high-precision adaptive control of the rotating bridge posture; The hierarchical warning and collaborative control module automatically matches the corresponding control strategies by establishing a multi-level warning mechanism and warning-control mapping matrix, thus achieving risk hierarchical response and intelligent collaborative control; The digital twin interactive verification module builds a high-precision digital twin model of the rotating bridge, and realizes virtual verification and dynamic simulation of the control strategy through virtual and real two-way data interaction and real-time comparative analysis.

2. A swivel bridge posture monitoring system based on dynamic feedback control according to claim 1, characterized in that: The multi-source sensor integration module includes the following components: Multi-type sensor unit, integrating GNSS / RTK, triaxial accelerometer, inclinometer, displacement sensor, fiber Bragg grating and strain gauge, to achieve comprehensive monitoring of bridge displacement, angle, acceleration and strain; High-precision time synchronization unit, using PTP / IEEE 1588 precise time protocol and distributed clock synchronization technology to ensure accurate and unified timestamps of sensor data at different locations; Energy management unit, combined with intelligent power management, ensures long-term stable operation of the sensor network in harsh environments; The network communication and redundant backup unit adopts a hybrid wired and wireless communication mode to achieve industrial-grade network transmission reliability, and ensures the continuity of data transmission through redundant communication path design; The sensor hardware self-diagnosis unit realizes real-time monitoring of the sensor hardware status and periodic self-calibration, and automatically identifies hardware-level physical faults to ensure the reliability of the data source; Environmental adaptability protection unit provides waterproof, dustproof, shockproof, anti-electromagnetic interference and temperature compensation protection measures for various sensors to ensure stable operation in harsh environments; The sensor layout optimization unit determines the optimal sensor layout plan for key parts of the bridge, so as to obtain the most comprehensive structural information with the least number of sensors.

3. The swivel bridge posture monitoring system based on dynamic feedback control according to claim 2 is characterized in that: The GNSS / RTK positioning accuracy in the multi-type sensor unit is better than 1 cm, the three-axis accelerometer range is ±16g and the sensitivity is not less than 0.001g, the inclination sensor measurement range is ±90° and the accuracy is better than 0.01°, the displacement sensor measurement range is 0-500 mm and the accuracy is better than 0.1 mm, and the fiber grating strain sensor measurement accuracy is better than 1 microstrain.

4. The swivel bridge posture monitoring system based on dynamic feedback control according to claim 1 is characterized in that: The edge computing and data fusion module includes the following components: Distributed edge computing units are deployed in key monitoring areas of bridges to enable on-site processing of sensor data; Data preprocessing and noise filtering unit, based on multi-scale Kalman filtering algorithm, handles the noise problem in the sensor raw data and improves the quality of single-source signals; The data quality assessment unit automatically identifies and repairs data-level outliers, missing values, and mutation values ​​in sensor data through statistical learning and pattern recognition methods to ensure the reliability of data fusion; The computing resource dynamic scheduling unit intelligently allocates and schedules edge computing resources based on the monitoring task priority and real-time data traffic to ensure the real-time and overall efficiency of key data processing; The multi-source data fusion unit dynamically selects appropriate data fusion algorithms for different monitoring scenarios to achieve comprehensive analysis and information extraction of multi-source heterogeneous data; The edge-cloud data distribution unit establishes a layered computing architecture and data transmission mechanism between edge nodes and cloud platforms. Lightweight real-time tasks are completed at the edge, and complex analysis tasks are pushed to the cloud. The data compression and transmission optimization unit adopts a context-aware compression algorithm to dynamically adjust the compression ratio according to data importance and bandwidth conditions, thereby optimizing data transmission efficiency while ensuring monitoring accuracy.

5. The swivel bridge posture monitoring system based on dynamic feedback control according to claim 1 is characterized in that: The Multiphysics Coupled Modeling Module includes the following components: Structural mechanics modeling unit, which builds a refined finite element model of the rotating bridge and calculates the stress distribution and structural deformation under various loads; Temperature field coupling unit simulates the non-uniform thermal expansion effects of sunlight radiation and ambient temperature changes on bridge components, and calculates the impact of thermal stress and thermal deformation on bridge posture; Contact friction mechanics unit, accurately modeling the friction characteristics and wear state of key contact interfaces, and calculating the damping effect of friction torque on the rotation process; Fluid-structure interaction unit, which simulates the dynamic effects of wind loads and water flow on bridges and predicts attitude fluctuations by calculating fluid-induced vibrations and aerodynamic effects; The physical model boundary condition unit continuously calibrates the model's physical boundary constraints and load conditions based on sensor monitoring data to ensure the model's physical consistency; The multi-scale computing collaborative unit integrates the multi-scale analysis technology of macroscopic structural response and microscopic material behavior to improve the physical accuracy of the model while ensuring computational efficiency.

6. The swivel bridge posture monitoring system based on dynamic feedback control according to claim 5 is characterized in that: The temperature field coupling unit is configured to achieve: ambient temperature monitoring range -40℃ to 85℃, with an accuracy better than ±0.5℃; solar radiation intensity monitoring range 0-1200W / m², with an accuracy better than ±5%; support 24-hour temperature field evolution prediction, with a prediction error less than ±2.5℃; thermal deformation calculation accuracy better than ±5% of the measured value; automatic identification of high-risk areas with temperature gradients exceeding 15℃ / m, and give early warnings.

7. The swivel bridge posture monitoring system based on dynamic feedback control according to claim 1 is characterized in that: The adaptive control strategy module includes the following components: The prediction control unit predicts the future state response of the bridge based on the multi-physics field coupling model, calculates the optimal control instruction sequence through rolling time domain optimization, and realizes feedforward control of the rotation process; The control parameter online identification unit continuously updates the control-related dynamic parameters using real-time monitoring data, capturing changes in structural characteristics, friction coefficients, and external loads; Reinforcement learning optimization unit, extracting the best strategy from historical control experience, adapting to complex environmental changes and model uncertainty, and continuously improving control effects; The control scheme optimization unit balances the rotation accuracy, energy consumption, execution time and safety margin targets, and generates the control scheme with the best comprehensive performance through the Pareto optimal method; Robustness assurance unit, which designs control laws with anti-interference capabilities against parameter disturbances and external interference, ensuring control stability in harsh environments and when some sensors fail; Actuator distribution unit, based on the characteristics and status of multiple bridge actuators, optimizes the distribution of control forces to avoid actuator overload and structural stress concentration; The control strategy performance evaluation unit conducts real-time evaluation and quantitative analysis of the implementation effect of the control strategy, providing a basis for the iterative optimization of the control strategy.

8. The swivel bridge posture monitoring system based on dynamic feedback control according to claim 1 is characterized in that: The hierarchical warning and collaborative control module includes the following components: Multi-dimensional status assessment unit, which comprehensively analyzes structural stress, displacement, vibration and environmental condition parameters, and calculates bridge health status indicators in real time; The warning level classification unit establishes a five-level warning threshold system of normal, attention, warning, danger and emergency based on bridge safety standards and historical data, realizing accurate classification of abnormal conditions; The early warning-control mapping matrix unit establishes a dynamic mapping relationship between early warning levels and control strategies, automatically matching corresponding control parameters and execution plans for different levels of risks; The multi-system coordination unit coordinates the linkage response of the swivel mechanism, shock absorption system and locking device to ensure that the overall system can coordinate to deal with abnormal conditions of different levels; The human-machine interactive confirmation unit provides manual intervention and decision confirmation mechanisms for high-level warning events, combining operator experience with automatic control to form a double guarantee; The emergency downgrade control unit automatically starts the downgraded operation mode in case of partial failure of the sensor or controller, maintains the basic control functions and ensures the safety of the swing bridge; The early warning response recording unit records the triggering conditions of the early warning events, the system response process and the event processing results, providing data support for the subsequent improvement of the early warning mechanism.

9. The swivel bridge posture monitoring system based on dynamic feedback control according to claim 1, characterized in that: The digital twin interaction verification module includes the following components: High-precision geometric modeling unit, based on laser scanning and BIM technology, builds a digital twin model of the rotating bridge, achieving accurate mapping from physical structure to virtual model; Integrated physical environment simulation unit, based on multi-physics field model, builds a comprehensive simulation environment including mechanics, thermals, and fluids as the basic platform for virtual testing and verification; The physical-virtual synchronization unit realizes the real-time mapping of monitoring data to the virtual model through a dedicated data interface, maintaining the dynamic consistency between the digital twin and the physical bridge status; The control strategy virtual verification unit uses the digital twin environment to pre-test the control strategy, evaluate its safety and effectiveness, and provide suggestions for the optimization of the actual control solution; Virtual-real difference analysis unit, which compares the deviation between the digital twin prediction results and the actual monitoring data, identifies model defects and potential anomalies, and continuously optimizes the accuracy of the digital twin model; Multi-scenario rehearsal unit simulates the rotation behavior of bridges under special working conditions, formulates emergency plans in advance, and enhances the ability to deal with complex situations; The visual interaction unit intuitively displays the bridge status and control process through a three-dimensional visual interface and augmented reality technology, supporting engineers in immersive analysis and decision-making assistance.

10. A rotating bridge posture monitoring system based on dynamic feedback control according to claim 9, characterized in that: The digital twin model constructed by the high-precision geometric modeling unit has the following characteristics: The geometric accuracy of key parts is better than ±5mm, and the overall geometric accuracy is better than ±20mm; Contains no less than 300 independent components and 60 key connection nodes; The total model data volume does not exceed 5GB to ensure smooth operation on ordinary workstations; Support at least 3 standard data exchange formats, including IFC, STEP and STL; Adopting parametric modeling technology, it supports rapid model update, and the response time for single parameter modification is less than 10 seconds; The data synchronization delay with the BIM system is less than 5 minutes; Supports dynamic switching of three levels of detail (LOD200-LOD400) to balance display accuracy and operating efficiency.

Images