Multi-mode intelligent cooperative suppression method and system for wind vibration during construction period of cable-stayed bridge

Abstract

The invention provides a multi-mode intelligent cooperative suppression method and system for wind vibration in a cable-stayed bridge construction period, and relates to the technical field of cooperative control, and the method comprises the steps: mapping three-dimensional wind speed vector field data and a strain field to a unified three-dimensional space-time grid coordinate system, and carrying out the screening to generate a composite risk focus coordinate set; defining a weighted Voronoi subdivision weight according to main beam rigidity distribution and path probability density, and generating a responsibility partition set; establishing a damping force-current-vibration response empirical mapping model, and solving an optimal current sequence through model prediction control; bridge floor acceleration response data after vibration suppression are monitored through an MEMS accelerometer array, and an optimized control strategy is formed; model parameters are updated online by using MAML meta-learning, and cross-section migration deployment of the model is realized through lightweight adjustment, so that multi-mode intelligent collaborative suppression of wind vibration during the construction period of the cable-stayed bridge is completed. According to the method, multi-mode intelligent cooperative suppression of wind vibration in the construction period of the cable-stayed bridge is achieved, and the stability and safety of a bridge structure are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of cooperative control, and in particular, to a multi-modal intelligent cooperative suppression method and system for wind-induced vibration during the construction period of a cable-stayed bridge. Background Art

[0002] In recent years, with the rapid development of the construction of long-span cable-stayed bridges, the problem of wind-induced vibration control during the construction period has become increasingly prominent. Traditional vibration suppression methods mainly rely on passive dampers or temporary counterweights. Although they can relieve some vibrations, they have significant limitations. For example, during the construction period, the bridge structure is vulnerable to wind-induced vibration, leading to structural safety and construction progress problems.

[0003] For the existing wind-induced vibration control technologies, on the one hand, passive control cannot adapt to the time-varying structural stiffness and transient wind field characteristics during the construction period, resulting in low vibration suppression efficiency; on the other hand, the existing technologies mostly adopt a single-modal control strategy, which is difficult to cope with the common multi-modal coupled vibration problems in the construction of cable-stayed bridges. For example, in the method based on a fixed-parameter tuned mass damper in the existing technology, although it can suppress vibrations at specific frequencies, it cannot be dynamically adjusted to adapt to the stiffness changes during the cantilever erection process and is prone to failure in a strong turbulent wind field. In addition, the existing solutions lack systematic optimization of the cooperative control of multiple actuators, resulting in uneven resource allocation, insufficient vibration suppression in some areas and excessive energy consumption in other areas. Summary of the Invention

[0004] The purpose of the present invention is to provide a multi-modal intelligent cooperative suppression method and system for wind-induced vibration during the construction period of a cable-stayed bridge to improve the above problems. To achieve the above purpose, the technical solutions adopted by the present invention are as follows: In the first aspect, the present application provides a multi-modal intelligent cooperative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge, including: Using a lidar array to collect three-dimensional wind speed vector field data in the construction area, and simultaneously synchronously monitoring the strain field of the main girder joints through distributed fiber Bragg grating sensors. Through high-precision time synchronization technology, map the three-dimensional wind speed vector field data and the strain field to a unified three-dimensional space-time grid coordinate system, calculate the strain gradient amplitude based on spatial difference, and combine the local vorticity component of the wind speed field and a double-threshold criterion to screen and generate a set of composite risk focus coordinates, denoted as the target area for damper dynamic deployment; Construct a bionic pheromone concentration field based on the set of composite risk focus coordinates, set update rules, generate the path probability density distribution of the damper in combination with the strain gradient amplitude, define the weighted Voronoi partition weight according to the main girder stiffness distribution and the path probability density, generate a set of responsibility partitions, verify the standard deviation of the vortex-induced vibration energy in each partition, and if not satisfied, iteratively adjust the partition weight until it meets the standard; The local average wind speed is obtained by processing the three-dimensional wind speed vector field data in space and time, and combined with the characteristic height of the main girder, the target intervention frequency is inversely deduced through the Strouhal number criterion; according to the responsibility partition set, a micro-aerodynamic fin array is deployed inside the composite risk focus coordinate set. Based on the target intervention frequency, a real-time angle of attack command is generated, and an empirical mapping model of damping force-current-vibration response is established. The optimal current sequence is solved through model predictive control; based on the time slot allocation protocol of the time-sensitive network, control commands are issued, where the control commands are to send the angle of attack command and the current spectrum to the edge computing node, and then control the aerodynamic fins and the magnetorheological damper; According to the control command, the acceleration response data of the bridge deck after vibration suppression is monitored by the MEMS accelerometer array, the energy decay rate is calculated, and accordingly, the angle of attack of the aerodynamic fin is adjusted. At the same time, the digital twin technology is used to generate a predicted response, and the control parameters are dynamically adjusted to form an optimized control strategy; The optimized control strategy is adopted to classify and store the historical data according to the wind field characteristic parameters, and a multi-modal data set is constructed; the MAML meta-learning is used to update the model parameters online. For the new construction section, the matching wind field parameters are loaded to initialize the meta-model, and the cross-section migration deployment of the model is realized through lightweight adjustment, so as to complete the multi-modal intelligent collaborative suppression of the wind vibration during the construction period of the cable-stayed bridge.

[0005] Preferably, the three-dimensional wind speed vector field data of the construction area is collected by the lidar array, and at the same time, the strain field of the main girder joint is synchronously monitored by the distributed fiber Bragg grating sensor. Through the high-precision time synchronization technology, the three-dimensional wind speed vector field data and the strain field are mapped to a unified three-dimensional space-time grid coordinate system. Based on the spatial difference calculation, the strain gradient amplitude is obtained, and combined with the local vorticity component of the wind speed field and the double-threshold criterion, the composite risk focus coordinate set is screened and generated, which is recorded as the target area for the dynamic deployment of the damper, including: The three-dimensional wind speed vector field data of the construction area is obtained by the lidar array with a spatial resolution of 0.5 m and a time sampling interval of 0.1 s, and the strain field of the main girder joint is monitored in real time by the distributed fiber Bragg grating sensor network with a sensitivity of 1 microstrain and a sampling frequency of 200 Hz; the carrier phase differential technology and the 1588 time protocol are used to map the data of the three-dimensional wind speed vector field data and the strain field to a unified three-dimensional space-time grid coordinate system to generate a fusion data set; Perform spatial difference operations on the strain field, calculate the strain gradient vector at each spatial point of the strain field at the main girder joint, and extract the Euclidean norm in the strain gradient vector as the gradient amplitude. Determine whether the gradient amplitude exceeds 50 με / m. If it exceeds, mark it as a potential stress concentration area; if it does not exceed, do not mark it. Calculate the local vorticity component in the three-dimensional wind speed vector field data, set a threshold to determine whether the local vorticity component is greater than the threshold. If it exceeds, mark it as a strong rotation wind field area; if it does not exceed, do not mark it. Traverse all spatio-temporal grid points, filter out the points that simultaneously include the potential stress concentration area and the strong rotation wind field area, and determine the grid points that meet the conditions as the high-risk points of wind-structure coupling. Integrate the grid points of all wind-structure coupling high-risk points to generate a composite risk focus coordinate set, which is recorded as the priority target area for damper dynamic deployment and aerodynamic fin regulation.

[0006] Preferably, construct a bionic pheromone concentration field based on the composite risk focus coordinate set, set an update rule, generate the path probability density distribution of the damper in combination with the strain gradient amplitude, define the weighted Voronoi tessellation weight according to the main girder stiffness distribution and the path probability density, generate a set of responsibility partitions, and verify the standard deviation of the vortex-induced vibration energy in each partition. If it does not meet the requirements, iteratively adjust the tessellation weight until it meets the standard, including: In the composite risk focus coordinate set, assign an initial pheromone concentration value to all coordinate points, construct a bionic pheromone concentration field using the Gaussian kernel density estimation method, and perform dynamic update using a dual-time scale mechanism suitable for the cantilever construction of cable-stayed bridges, so as to obtain the updated pheromone concentration field. The dual-time scale mechanism includes short-time scale update and long-time scale reset; According to the updated pheromone concentration field and the strain gradient amplitude, set a dynamic update rule and calculate the path probability density of the damper; Combine the path probability density of the damper and the time-varying model of the main girder stiffness, apply the weighted Voronoi tessellation algorithm to generate a set of responsibility partitions, use the MEMS accelerometer array to collect data at a sampling rate of 200 Hz, and calculate the standard deviation of the acceleration response in each partition. If the standard deviation of any partition exceeds 8%, adjust the tessellation weight and re-tessellate until the standard deviation of the acceleration response in all partitions meets the standard, so as to obtain a set of partitions that meet the standards.

[0007] Preferably, based on the set of responsibility partitions, deploy a micro-aerodynamic fin array inside the composite risk focus coordinate set, generate a real-time angle of attack command based on the target intervention frequency, and establish an empirical mapping model of damping force-current-vibration response, and solve the optimal current sequence through model predictive control, including: According to the risk assessment in the set of responsibility partitions, deploy a micro-aerodynamic fin array in the composite risk focus coordinate set in the form of an equidistant matrix; Calculate the real-time angle of attack of each pneumatic flap at time using the target intervention frequency t of; Based on the relationship between the damping force, current, and vibration response, construct an empirical mapping model and obtain the optimal current sequence through model predictive control.

[0008] Preferably, according to the control instruction, monitor the acceleration response data of the bridge deck after vibration suppression through the MEMS accelerometer array, calculate the energy decay rate, and accordingly adjust the angle of attack of the pneumatic flap. At the same time, use digital twin technology to generate a predicted response and dynamically adjust the control parameters to form an optimized control strategy, including: Deploy the MEMS accelerometer array at key positions of the bridge according to the control instruction, where the key positions include main girders, bridge towers, and stay cable structural components, and record the acceleration data of each sensor; Analyze the collected acceleration data and calculate the energy decay rate by comparing the energy changes of the system before and after vibration suppression; Based on the energy decay rate, adjust the angle of attack of the pneumatic flap to obtain the adjusted angle of attack direction; Use the collected real-time wind speed vector field data and bridge structure response data for simulation in the digital twin model to predict the response of the bridge structure in the future for a period of time. Compare the predicted response obtained from the simulation with the actual monitoring data. If there are differences, optimize them, and combine the optimization results with the adjusted angle of attack direction to evaluate the current control strategy; Dynamically adjust the control parameters in response to the energy decay rate and the current control strategy, where adjusting the control parameters includes iterative solution of model predictive control, optimizing the pneumatic flap angle of attack adjustment strategy, and the magnetorheological damper current control sequence, so as to achieve the collaborative suppression of wind vibration and adapt to the current wind field conditions and bridge structure response.

[0009] In a second aspect, the present application also provides a multi-modal intelligent collaborative wind vibration suppression system for cable-stayed bridges during construction, including: Mapping calculation module: used to collect three-dimensional wind speed vector field data of the construction area using a lidar array, and at the same time synchronously monitor the strain field of the main girder joint through a distributed fiber optic grating sensor. Through high-precision time synchronization technology, map the three-dimensional wind speed vector field data and the strain field to a unified three-dimensional space-time grid coordinate system, calculate the strain gradient amplitude based on spatial difference, and combine the local vorticity component of the wind speed field and the double-threshold criterion to screen and generate a set of composite risk focus coordinates, denoted as the target area for damper dynamic deployment; Generation module: It is used to construct a bionic pheromone concentration field based on the composite risk focus coordinate set, set the update rule, generate the path probability density distribution of the damper by combining the strain gradient amplitude, define the weighted Voronoi tessellation weight according to the main girder stiffness distribution and the path probability density, generate the responsibility partition set, verify the standard deviation of the vortex-induced vibration energy in each partition, and if not satisfied, iteratively adjust the tessellation weight until it meets the standard; Establishment module: It is used to obtain the local average wind speed by processing the three-dimensional wind speed vector field data in space and time, and inversely deduce the target intervention frequency through the Strouhal number criterion in combination with the characteristic height of the main girder; According to the responsibility partition set, deploy a micro-aerodynamic flap array inside the composite risk focus coordinate set, generate real-time angle-of-attack commands based on the target intervention frequency, and establish an empirical mapping model of damping force-current-vibration response, and solve the optimal current sequence through model predictive control; Based on the time-sensitive network's time slot allocation protocol, send control commands, where the control commands are to send the angle-of-attack command and the current spectrum to the edge computing node, and then control the aerodynamic flaps and magnetorheological dampers; Adjustment and optimization module: It is used to monitor the bridge deck acceleration response data after vibration suppression through the MEMS accelerometer array according to the control command, calculate the energy decay rate, and adjust the angle of attack of the aerodynamic flap accordingly. At the same time, use digital twin technology to generate predicted responses, dynamically adjust control parameters, and form an optimized control strategy; Construction module: It is used to adopt the optimized control strategy, classify and store historical data according to the wind field characteristic parameters, and construct a multi-modal data set; Use MAML meta-learning to update the model parameters online, initialize the meta-model by loading the matching wind field parameters for the new construction section, and achieve cross-section migration and deployment of the model through lightweight adjustment, so as to complete the multi-modal intelligent collaborative suppression of wind-induced vibration during the construction period of the cable-stayed bridge.

[0010] Thirdly, the present application also provides a multi-modal intelligent collaborative suppression device for wind-induced vibration during the construction period of a cable-stayed bridge, including: A memory for storing computer programs; A processor for implementing the steps of the multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of the cable-stayed bridge when executing the computer program.

[0011] Fourthly, the present application also provides a readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the above-mentioned multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge.

[0012] The beneficial effects of the present invention are: The present invention utilizes a lidar array and distributed fiber Bragg grating sensors to collect wind speed vector field and strain field data, maps them to a unified coordinate system through high-precision time synchronization technology, calculates the magnitude of the strain gradient, and combines vorticity components and double thresholds to screen out the risk focus coordinate set; constructs a bionic pheromone concentration field based on the risk focus coordinate set, generates the path probability density distribution of the damper in combination with the magnitude of the strain gradient, generates a responsibility partition set through weighted Voronoi tessellation, and verifies the standard deviation of the vortex-induced vibration energy.

[0013] According to the responsibility partition set, the present invention dynamically adjusts the attack angle of the micro pneumatic flap and the current of the magnetorheological damper, solves the optimal current sequence through model predictive control, realizes the continuous and effective suppression of wind vibration, generates a predicted response using digital twin technology, compares it with the actual monitoring data, evaluates the effect of the control strategy, and provides a basis for adjusting the control parameters. Based on the optimized control strategy, a multi-modal data set is constructed, the model parameters are updated online using MAML meta-learning, and cross-section migration deployment of the model is achieved through lightweight adjustment, completing the multi-modal intelligent collaborative suppression of wind vibration during the construction period of the cable-stayed bridge.

[0014] Other features and advantages of the present invention will be described in the subsequent specification, and, in part, will be obvious from the specification, or will be understood by implementing the embodiments of the present invention. The objectives and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in the written specification, claims, and drawings. Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other relevant drawings can also be obtained based on these drawings.

[0016] Figure 1 It is a schematic flow chart of the multi-modal intelligent collaborative suppression method for wind vibration during the construction period of the cable-stayed bridge described in the embodiments of the present invention; Figure 2 It is a schematic structural diagram of the multi-modal intelligent collaborative suppression system for wind vibration during the construction period of the cable-stayed bridge described in the embodiments of the present invention; Figure 3 It is a schematic structural diagram of the multi-modal intelligent collaborative suppression equipment for wind vibration during the construction period of the cable-stayed bridge described in the embodiments of the present invention.

[0017] In the figure: 701, mapping calculation module; 702, generation module; 703, establishment module; 704, adjustment and optimization module; 705, construction module; 800, multi-modal intelligent collaborative suppression device for wind-induced vibration during the construction period of cable-stayed bridge; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, communication component. Detailed implementation mode

[0018] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but is merely representative of selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not require further definition and explanation in subsequent drawings. At the same time, in the description of the present invention, terms such as "first" and "second" are only used for differential description and cannot be construed as indicating or implying relative importance.

[0020] Embodiment 1:

[0021] This embodiment provides a multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge.

[0022] See Figure 1 , the figure shows that this method includes step S100, step S200, step S300, step S400 and step S500.

[0023] S100. Use a lidar array to collect three-dimensional wind speed vector field data in the construction area, and simultaneously monitor the strain field of the main girder joint through a distributed fiber Bragg grating sensor. Through high-precision time synchronization technology, map the three-dimensional wind speed vector field data and the strain field to a unified three-dimensional space-time grid coordinate system. Based on spatial difference calculation, obtain the strain gradient amplitude, and combine the local vorticity component of the wind speed field and the double-threshold criterion to screen and generate a composite risk focus coordinate set, denoted as the target area for damper dynamic deployment.

[0024] It can be understood that in this step S100, it includes S101, S102 and S103, where: S101. Obtain the three-dimensional wind speed vector field data of the construction area by using a lidar array with a spatial resolution of 0.5 m and a time sampling interval of 0.1 s, and real-time monitor the strain field of the main girder joint through a distributed fiber Bragg grating sensor network with a sensitivity of 1 microstrain and a sampling frequency of 200 Hz; use carrier phase differential technology and the 1588 time protocol to map the three-dimensional wind speed vector field data and the strain field data to a unified three-dimensional spatio-temporal grid coordinate system to generate a fusion data set; It should be noted that a lidar array is used to obtain the three-dimensional wind speed vector field data of the construction area, and the three-dimensional wind speed vector field data includes wind speed components , and , a spatial resolution of 0.5 m and a time sampling interval of 0.1 s, and real-time monitor the strain field of the main girder joint through a distributed fiber Bragg grating sensor network , a sensitivity of 1 microstrain and a sampling frequency of 200 Hz. Set up a Beidou RTK reference station in the bridge construction area to provide real-time differential positioning signals for the lidar array and fiber Bragg grating sensor nodes (roving stations), and correct the spatial coordinates of each sensor to a unified construction coordinate system (such as the center of the bridge tower as the origin), with a planar positioning error ≤ 1 cm and an elevation error ≤ 2 cm; the lidar point positions obtain absolute coordinates through RTK , and the coordinates of the fiber sensor nodes are marked as . Deploy an IEEE 1588 Precision Time Protocol (PTP) master clock, and distribute synchronization signals to all sensor nodes through an optical fiber network to achieve full-system clock synchronization, with a time jitter < 100 nanoseconds, and the timestamp of the lidar wind speed data is , and the timestamp of the fiber strain data is , and the absolute value of the difference between the timestamp of the lidar wind speed data and the time of the fiber strain data is less than 1 microsecond. Then, with the longitudinal direction of the main girder as the X-axis, the transverse direction as the Y-axis, and the vertical direction as the Z-axis, construct a spatial grid: among them, the grid resolution covering the main girder and the cable tower construction area is , and the maximum vortex-induced vibration frequency corresponding to the time step is , where the grid point coordinates are , where , j , k is the spatial index, m is the time index. Then perform data interpolation and fusion, where the lidar discrete point cloud data is mapped to the grid through inverse distance weighted interpolation, and the fiber sensor data is assigned to the grid based on the nearest neighbor principle. If there is a sensor node p around r =0.2 m inside q , then , where, is the mapped strain value at the grid point , and is the measured strain value at the physical point . Otherwise, linear interpolation is used for filling, and the wind speed and strain data are aligned according to the time step to generate a fused dataset , where , and are the spatial coordinates (longitudinal, transverse, and vertical directions along the bridge), is the time series point, , and are the three-dimensional wind speed components, is the strain value. Finally, the residual between the interpolated data and the original data is calculated , where δ is the difference norm between the grid interpolation data and the original data, is the data matrix after gridification, is the original sensor data matrix. If δ the maximum value of is greater than the threshold, such as the wind speed residual > 0.2 m / s or the strain residual > 5 με), then local grid refinement (resolution is increased to 0.2 m) is triggered, and at tm time, the gradient mutation of adjacent grid points is checked (such as ∥ ∇ v∥ > 3 m / s / m or ∥ ∇ ε∥ > 100 με / m ), then the abnormal area is marked and the sensor status is manually reviewed.

[0025] S102. Perform spatial difference operations on the strain field, calculate the strain gradient vector of each spatial point in the strain field of the main girder joint, and extract the Euclidean norm in the strain gradient vector as the gradient amplitude. Determine whether the gradient amplitude exceeds 50 με / m. If it exceeds, it is marked as a potential stress concentration area; if it does not exceed, it is not marked; calculate the local vorticity component in the three-dimensional wind speed vector field data, set a threshold to determine whether the local vorticity component is greater than the threshold. If it exceeds, it is marked as a strong rotating wind field area; if it does not exceed, it is not marked; S103. Traverse all spatio-temporal grid points , where is the coordinate of the m-th measurement point in the longitudinal direction of the bridge, $n$th measuring point coordinates in the transverse direction of the bridge, select points that simultaneously include potential stress concentration areas and strong rotating wind field areas, and determine the grid points that meet the conditions as high-risk points of wind-structure coupling. Integrate the grid points of all high-risk points of wind-structure coupling to generate a composite risk focus coordinate set, denoted as the priority target area for damper dynamic deployment and aerodynamic fin regulation.

[0026] It should be noted that for each spatial point, the partial derivatives of its strain in the x, y (and z direction if three-dimensional) are calculated by the central difference method to obtain the strain gradient vector of the point, that is, to identify the stress concentration area at the main girder joint (such as high-gradient areas such as the anchorage end and welding seam). The calculation object of the strain gradient vector obtained in this step is the spatial strain change rate of each discrete measurement point in the strain field, and the partial derivatives in each direction are directly obtained through the difference operation and combined into a vector.

[0027] In this step , the set threshold of the wind field vorticity criterion is: , screen the areas with significant rotation intensity in the local wind field. And the set threshold of the strain gradient amplitude is: identify the stress concentration risk area at the main girder joint, and screen according to the composite conditions. Screen the points that simultaneously meet the following conditions: , where is the vorticity in the z direction (perpendicular to the bridge deck direction), and then generate a composite risk focus coordinate set, that is , the output is used as the priority target area for damper dynamic deployment and aerodynamic fin regulation to ensure that the suppression resources accurately cover the strong coupling area of wind vibration energy and structural response, and at the same time balance the sensitivity and false alarm rate through the double-threshold value.

[0028] S200. Based on the composite risk focus coordinate set, construct a bionic pheromone concentration field, set update rules, generate the path probability density distribution of the damper in combination with the strain gradient amplitude, define the weighted Voronoi partition weight according to the main girder stiffness distribution and the path probability density, generate a set of responsibility partitions, and verify the standard deviation of the vortex vibration energy in each partition. If not satisfied, iterate and adjust the partition weight until it meets the standard.

[0029] It can be understood that in this step S200, it includes S201, S202 and S203, where: S201. In the composite risk focus coordinate set, assign an initial pheromone concentration value to all coordinate points, construct a bionic pheromone concentration field using the Gaussian kernel density estimation method, and perform dynamic update using a double-time scale mechanism suitable for the cantilever construction of cable-stayed bridges, so as to obtain the updated pheromone concentration field, where the double-time scale mechanism includes short-time scale update and long-time scale reset; It should be noted that initial values are assigned to all coordinate points in the composite risk focus coordinate set , and define the dynamic evolution formula of pheromone concentration as , where is the strain gradient amplitude, which inhibits the pheromone accumulation in high-stress areas and outputs the dynamic pheromone field characterizes the path attractiveness distribution of the risk area. In this step, the short-term scale is updated as: , where represents the increment of vortex-induced vibration energy, represents the strain gradient amplitude of the temporary anchorage area, is time t is the pheromone concentration field at time λ is the pheromone decay coefficient, is the short-term scale; long-term scale reset (for each assembled segment): When the main girder stiffness distribution K ( x , y ) changes by more than 15% due to segment assembly, re-initialize the concentration field.

[0030] S202. According to the updated pheromone concentration field and strain gradient amplitude, set the dynamic update rule and calculate the path probability density of the damper. The calculation formula is as follows:

[0031] In the formula, is the priority of deploying the damper at the coordinate ( x , y ), is the bionic pheromone concentration field, is the strain gradient amplitude, is the vorticity threshold function, is the threshold set for the cantilever construction of the steel box girder, is the global integral normalization factor, which integrates over the entire bridge deck area; In this step, this probability density reflects the layout priority of the damper at different positions during the wind vibration suppression of the cable-stayed bridge construction period, and it is preferentially arranged in high-risk areas (high vorticity and high strain gradient areas). Through normalization, ensure that the sum of the probability density on the bridge deck is 1, so as to provide a scientific basis for the optimal layout of the damper.

[0032] S203. Combine the path probability density of the damper and the time-varying model of the main girder stiffness, apply the weighted Voronoi tessellation algorithm to generate the set of responsibility partitions, use the MEMS accelerometer array to collect data at a sampling rate of 200 Hz, and calculate the standard deviation of the acceleration response within each partition; if the standard deviation of any partition exceeds 8%, adjust the tessellation weight and re-tessellate until the standard deviation of the acceleration response of all partitions meets the standard, so as to obtain a set of partitions that meet the standards.

[0033] It should be noted that in this step, the acquisition and calculation of the time-varying model of the main girder stiffness are derived from real-time monitoring data and computational analysis. The real-time monitoring data is obtained from the strain distribution and deformation data of the main girder acquired through a distributed optical fiber monitoring system. At the same time, the load information is obtained using a vehicle dynamic weighing system, and combined with the strain sensor data, the change in the main girder stiffness is evaluated. The computational analysis is based on the finite element method, considering the geometric shape, material properties, and initial prestress distribution of the main girder. By linearly fitting the monitoring data using the least squares method, a first-order linear equation for the change of the main girder stiffness over time is obtained, thereby obtaining the attenuation rate coefficient of the stiffness, and further obtaining an optimized time-varying model of the stiffness.

[0034] S300. The local average wind speed is obtained by processing the three-dimensional wind speed vector field data in space and time, and combined with the characteristic height of the main girder, the target intervention frequency is inversely deduced through the Strouhal number criterion. According to the responsibility partition set, a micro-aerodynamic fin array is deployed inside the compound risk focus coordinate set. Based on the target intervention frequency, a real-time angle of attack command is generated, and an empirical mapping model of damping force - current - vibration response is established. The optimal current sequence is solved through model predictive control. Based on the time-sensitive network's time slot allocation protocol, control commands are issued, where the control commands are to send the angle of attack command and the current spectrum to the edge computing node, thereby controlling the aerodynamic fins and the magnetorheological damper.

[0035] It should be noted that in this step, the aerodynamic fin is a small device installed on the bridge structure. By adjusting its angle of attack, the aerodynamic force distribution on the bridge surface is changed, thereby suppressing the vibration of the bridge under the action of wind. This method is particularly important in long-span cable-stayed bridges because such bridges are prone to vortex-induced vibration (VIV) under the action of wind, which affects the safety and service life of the bridge. The dynamic angle of attack regulation of the aerodynamic fin is achieved by real-time monitoring of the wind field and structural response, and generating a real-time angle of attack command according to the target intervention frequency to suppress the local aerodynamic elastic coupling effect. The current sequence is related to the magnetorheological damper. The magnetorheological damper is a smart damper whose damping force can be adjusted by changing the current. In the multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of cable-stayed bridges, an empirical mapping model of damping force - current - vibration response is established, and the optimal current sequence is solved using model predictive control to effectively control the vibration of the bridge. This method can accurately adjust the damping force to adapt to different wind field conditions and the dynamic response of the bridge. In summary, the aerodynamic fin and the current sequence play important roles in the multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of cable-stayed bridges. They act together on the bridge structure through intelligent control technologies to effectively suppress wind-induced vibration and ensure the safety and stability of the bridge.

[0036] It is understandable that the optimal current sequence, together with the angle of attack command of the pneumatic flap, is sent to the edge computing node based on the time-sensitive network (TSN) time slot allocation protocol to ensure that the control delay is less than 10 ms. After being sent, these commands will be used to control the pneumatic flap and the magnetorheological damper, so as to achieve intelligent collaborative suppression of the wind vibration of the cable-stayed bridge.

[0037] In this embodiment, the local average wind speed and the characteristic height of the main girder are important parameters for calculating the target intervention frequency and performing wind vibration control. The local average wind speed can be measured in the construction area by devices such as a lidar array, while the characteristic height of the main girder D generally refers to a representative dimension of the bridge main girder, such as height or width. The acquisition of these parameters is crucial for understanding and predicting hydrodynamic phenomena, such as wind load analysis on buildings and bridges. After obtaining the local average wind speed and the characteristic height of the main girder, the Strouhal number can be used to calculate the target intervention frequency. The Strouhal number is a dimensionless number used to describe the frequency of vortex shedding in the wake of an object in fluid flow, which relates the vortex shedding frequency to the flow velocity and the characteristic length of the object. As for the optimal current sequence, it is obtained through the nonlinear MPC control of the magnetorheological damper and is used to adjust the damping force to suppress the bridge vibration. This sequence, together with the angle of attack command of the pneumatic flap, is sent to the edge computing node through the time-sensitive network (TSN) time slot allocation protocol to ensure that the control delay is less than 10 ms. After sending these commands, the edge computing node will control the pneumatic flap and the magnetorheological damper, so as to achieve intelligent collaborative suppression of the wind vibration of the cable-stayed bridge. This process involves precise control strategies to adapt to different wind field conditions and the dynamic response of the bridge.

[0038] Among them, the three-dimensional wind speed vector field data usually includes the magnitude and direction information of the wind speed, which can be used to calculate the local average wind speed. The local average wind speed refers to the average value of the wind speed within a specific area or time period, and this parameter is crucial for understanding and predicting the impact of the wind on structures (such as cable-stayed bridges). In practical applications, the local average wind speed can be obtained by performing spatial and temporal averaging on the wind speed vector field data.

[0039] Among them, the local average wind speed can be obtained by statistically analyzing the collected wind speed data within a certain time and space range. For example, multiple wind speed sensors can be deployed in the bridge construction area to collect wind speed data at each location, and then these data are averaged to obtain the local average wind speed. This method can provide in-depth understanding of the wind field characteristics in the construction area, thus providing important basis for bridge design and construction. As for the characteristic height of the main girder D, this is a parameter related to bridge design, usually given in the design drawings or specifications of the bridge, which represents the characteristic dimensions such as the height or width of the main girder of the bridge. In wind engineering analysis, this parameter is used to calculate wind load related indicators such as the Strouhal number.

[0040] It can be understood that in this step S300, it includes S301, S302 and S303, where: S301. According to the risk assessment within the responsibility partition set, deploy the micro-aerodynamic flap array in the form of an equidistant matrix within the compound risk focus coordinate set; It can be understood that in this step, according to the risk assessment within the responsibility partition set, deploy the micro-aerodynamic flap array in the form of an equidistant matrix within the compound risk focus coordinate set. This deployment method can ensure that the flap array can cover the entire risk area without interference, thereby effectively controlling wind vibration, and adjusting the air flow in a timely manner when wind vibration occurs to reduce the impact on the bridge structure.

[0041] S302. Use the target intervention frequency to calculate the real-time angle of attack of each aerodynamic flap at time t The calculation formula is as follows:

[0042] In the formula, is the real-time angle of attack of the th aerodynamic flap, is the set value of the initial angle of attack, is the target intervention frequency, t is time, is the proportional control gain, is the strain value measured at position and time t , is the Heaviside step function, is the magnitude of the strain gradient, is the threshold of the strain gradient for triggering the adjustment of the angle of attack; the technical effect of this step is that it can dynamically adjust the angle of attack of the aerodynamic flap according to the real-time wind field and structural response, thereby effectively suppressing wind vibration.

[0043] S303. Based on the relationship between the damping force, current and vibration response, construct an empirical mapping model, and obtain the optimal current sequence through model predictive control. The calculation formula is as follows:

[0044] In the formula, is the current control sequence of the th magnetorheological damper, T is the total time range, that is, the upper limit of the time step, is the trajectory density distribution, is the energy consumption penalty coefficient, is time t of the current control sequence, is the structural vibration acceleration response, is the relationship among the damping force, current, and vibration response.

[0045] It should be noted that in this step, the optimal current sequence is solved through model predictive control, which can minimize energy consumption while ensuring the suppression effect, adapt to different wind field conditions and bridge structure responses, and improve the adaptability and robustness of the system. Solving the optimal current sequence through model predictive control can ensure that the magnetorheological damper can provide the best damping force under different wind field conditions, further enhancing the stability and safety of the bridge structure.

[0046] S400. According to the control instruction, monitor the acceleration response data of the bridge deck after vibration suppression through the MEMS accelerometer array, calculate the energy decay rate, and accordingly adjust the angle of attack of the aerodynamic fin. At the same time, use the digital twin technology to generate the predicted response and dynamically adjust the control parameters to form an optimized control strategy.

[0047] It can be understood that in this step S400, it includes S401, S402, S403, S404, and S405, where: S401. According to the control instruction, deploy the MEMS accelerometer array at the key positions of the bridge, where the key positions include the main girder, bridge tower, and stay cable structural components, and record the acceleration data of each sensor; It can be understood that through these sensors, the acceleration changes of the bridge under the influence of wind vibration are accurately monitored, the acceleration value of each sensor is recorded, and the actual dynamic behavior of the bridge structure under wind vibration is reflected.

[0048] S402. Analyze the collected acceleration data, and calculate the energy decay rate by comparing the energy changes of the system before and after vibration suppression. The calculation formula is as follows:

[0049] In the formula, is the energy decay rate at time t, is the accumulation of the energy changes of all monitoring points on the entire bridge structure, is the velocity response before vibration suppression, α is the strain energy weight coefficient, is the strain history data before vibration suppression, is the acceleration response after vibration suppression, is the real-time strain after vibration suppression, is to calculate the energy change from the initial moment to the current moment, is all the tiny time intervals; It should be noted that this formula calculates the change in the system energy before and after vibration suppression, and evaluates the effect of the vibration suppression measures by comparing the energies before and after vibration suppression. The energy decay rate η ( t ) The closer it is to 1, the better the vibration suppression effect, and the more the system energy decreases. This step can quantitatively evaluate the effectiveness of the vibration suppression measures and provide a basis for subsequent adjustment of the control strategy.

[0050] S403. Based on the energy decay rate, adjust the angle of attack of the pneumatic flap to obtain the adjusted angle of attack direction, and its calculation formula is as follows:

[0051] In the formula, is the gradient direction sign function, is the angle of attack of the th micro pneumatic flap at time t , is the position coordinate of the th micro pneumatic flap on the bridge structure, and t is the current time point; It should be noted that during the implementation of this method, based on the energy decay rate results calculated above, the angle of attack of the micro pneumatic flap will be adjusted to further optimize the wind vibration suppression effect. Specifically, the adjustment strategy of the angle of attack will be based on real-time monitoring data to dynamically change the angle of attack of the pneumatic flap to ensure that the direction of the counteracting force generated by the pneumatic flap is opposite to the direction of vorticity propagation, thereby effectively suppressing wind vibration.

[0052] In specific operations, the angle of attack of each pneumatic flap will be adjusted according to its position and the strain gradient at the current moment. If the strain gradient exceeds the preset threshold, it indicates that there may be a large stress concentration at this position. At this time, the angle of attack of the pneumatic flap will be increased or decreased accordingly to generate sufficient counteracting force to offset the vibration caused by wind vibration. In this way, the pneumatic flap can quickly respond to local wind field changes and achieve real-time suppression of the vibration of the bridge structure. That is to say, if the direction indicated by the gradient vector is the same as the current angle of attack direction, the angle of attack is increased by π radians (180 degrees) to change the aerodynamic direction of the pneumatic flap; if the direction is opposite, the angle of attack is decreased by π radians. This adjustment helps to optimize the performance of the pneumatic flap to suppress wind vibration.

[0053] The technical effect of this step is reflected in the ability to flexibly adjust the working state of the pneumatic fins according to the actual vibration state of the bridge structure and the wind field conditions. This can not only improve the adaptability and effectiveness of the wind vibration suppression measures, but also maintain the stability and safety of the structure during the entire bridge construction period. Through this dynamic adjustment strategy, the pneumatic fins can participate in the wind vibration control process more intelligently, providing key technical support for the multi-modal intelligent collaborative suppression of wind vibration during the bridge construction period.

[0054] S404. Use the collected real-time wind speed vector field data and bridge structure response data to conduct simulations in the digital twin model to predict the response of the bridge structure in the future for a period of time. Compare the predicted response obtained from the simulation with the actual monitoring data. If there are differences, optimize it, and combine the optimization results with the adjusted angle of attack direction to evaluate the current control strategy. It can be understood that in this step, using digital twin technology, the collected data is input into the digital twin model for simulation to predict the response of the bridge structure in the future for a period of time. The digital twin model can simulate the equipment model, upload the state data to the virtual model for simulation operation, run synchronously with the real equipment, simulate the future operation conditions, and solve the problems of high cost and poor information accuracy of the continuous operation of the actual equipment. Compare the predicted response obtained from the simulation with the actual monitoring data. This step involves comparing and analyzing the simulation results with the actual acceleration response data, and evaluating the effect of the current control strategy through quantitative analysis such as regression analysis and trend analysis. If there are significant differences between the predicted response and the actual monitoring data, it indicates that the current control strategy may need to be adjusted. According to the comparison results, evaluate the current control strategy and provide a basis for the subsequent adjustment of control parameters. This step involves continuously iteratively solving the model predictive control problem to achieve continuous and effective suppression of wind vibration. Its control strategy includes the angle of attack adjustment strategy of the pneumatic fins and the current control sequence of the magnetorheological damper, improving the adaptability and accuracy of the control strategy, and providing key technical support for the multi-modal intelligent collaborative suppression of wind vibration during the bridge construction period.

[0055] S405. Dynamically adjust the control parameters in response to the energy decay rate and the current control strategy, where adjusting the control parameters includes iteratively solving the model predictive control, optimizing the angle of attack adjustment strategy of the pneumatic fins and the current control sequence of the magnetorheological damper, so as to achieve the collaborative suppression of wind vibration and adapt to the current wind field conditions and bridge structure response.

[0056] It should be noted that through the above steps, the multi-modal intelligent collaborative suppression of wind vibration during the construction period of the cable-stayed bridge is achieved, ensuring the stability and safety of the bridge structure. It involves continuously iteratively solving the model predictive control problem to ensure that the control strategy adapts to the changing wind field conditions and bridge responses, and achieving the optimal wind vibration suppression effect.

[0057] S500. Adopt an optimized control strategy, classify and store historical data according to wind field characteristic parameters to construct a multi-modal data set; use MAML meta-learning to update model parameters online, load matching wind field parameters for a new construction section to initialize the meta-model, and achieve cross-section migration and deployment of the model through lightweight adjustment, so as to complete the multi-modal intelligent collaborative suppression of wind-induced vibration during the construction period of the cable-stayed bridge.

[0058] It can be understood that in this step, an optimized control strategy is adopted, and historical data is classified and stored according to wind field characteristic parameters (such as Froude number Fr and Reynolds number Re ) to construct a multi-modal data set, using the predicted response generated in the above step and the calculated energy decay rate to provide a data basis for subsequent model parameter updates.

[0059] It should be noted that in step S500, it includes S501, S502 and S503, where: S501. Adopt an optimized control strategy, classify and store historical data according to wind field characteristic parameters (such as Froude number Fr and Reynolds number Re ) to store the classified data as a multi-modal data set for subsequent model training and verification, using the predicted response generated in the above step and the calculated energy decay rate to provide a data basis for subsequent model parameter updates.

[0060] S502. Use the multi-modal data set to update model parameters online through the Model-Agnostic Meta-Learning (MAML) algorithm. For a new construction section, load matching wind field parameters to initialize the meta-model, and update the trajectory generator parameters through few-shot incremental learning to adapt to the new wind field conditions.

[0061] Among them, MAML is a meta-learning method that updates the parameters of the trajectory generator through few-shot incremental learning, thereby improving the generalization ability of the model. This step uses the multi-modal data set obtained above to update model parameters through online learning to adapt to the new wind field conditions.

[0062] S503. Adopt lightweight technology to adjust the meta-model, reduce model complexity and computational load, and deploy the adjusted model to a new construction section to achieve cross-section model migration and application.

[0063] It can be understood that lightweight is a technology that reduces model complexity and computational load, which helps the model to migrate and deploy between different sections, ensuring that the control strategy can adapt to the current wind field conditions and bridge structure responses, and achieving wind-induced vibration suppression.

[0064] Therefore, through the above steps, the multi-modal intelligent collaborative suppression of wind-induced vibration during the construction period of the cable-stayed bridge is completed. Not only historical data and current control strategies are used to optimize control parameters, but also through online learning and lightweight adjustment, the rapid adaptation and deployment of the model are realized, thereby improving the effect and efficiency of wind-induced vibration suppression.

[0065] Embodiment 2:

[0066] As Figure 2 shown, this embodiment provides a multi-modal intelligent collaborative suppression system for wind-induced vibration during the construction period of a cable-stayed bridge. Refer to Figure 2 The system includes: Mapping calculation module 701: It is used to collect three-dimensional wind speed vector field data of the construction area by using a lidar array, and at the same time synchronously monitor the strain field of the main girder joints through distributed fiber Bragg grating sensors. Through high-precision time synchronization technology, the three-dimensional wind speed vector field data and the strain field are mapped to a unified three-dimensional space-time grid coordinate system. Based on spatial difference calculation, the strain gradient amplitude is obtained, and combined with the local vorticity component of the wind speed field and the double-threshold criterion, a set of composite risk focus coordinates is screened and generated, which is recorded as the target area for damper dynamic deployment; Generation module 702: It is used to construct a bionic pheromone concentration field based on the set of composite risk focus coordinates, set update rules, generate the path probability density distribution of the damper in combination with the strain gradient amplitude, define the weighted Voronoi partition weight according to the main girder stiffness distribution and the path probability density, generate a set of responsibility partitions, verify the standard deviation of the vortex-induced vibration energy in each partition, and if not satisfied, iteratively adjust the partition weight until it meets the standard; Establishment module 703: It is used to obtain the local average wind speed by processing the three-dimensional wind speed vector field data in space and time, and combined with the characteristic height of the main girder, inversely deduce the target intervention frequency through the Strouhal number criterion; according to the set of responsibility partitions, deploy a micro-aerodynamic wing array inside the set of composite risk focus coordinates, generate real-time angle-of-attack instructions based on the target intervention frequency, and establish an empirical mapping model of damping force - current - vibration response, and solve the optimal current sequence through model predictive control; based on the time-slot allocation protocol of the time-sensitive network, send control instructions, where the control instructions are to send the angle-of-attack instructions and current spectra to the edge computing node, and then control the aerodynamic wing and the magnetorheological damper; Adjustment and optimization module 704: It is used to, according to the control instructions, monitor the bridge deck acceleration response data after vibration suppression through the MEMS accelerometer array, calculate the energy decay rate, and accordingly adjust the angle of attack of the aerodynamic wing. At the same time, use digital twin technology to generate predicted responses and dynamically adjust control parameters to form an optimized control strategy; Building block 705: It is used to classify and store historical data according to the wind field characteristic parameters by adopting an optimized control strategy, and construct a multi-modal data set; use MAML meta-learning to update the model parameters online, load the matching wind field parameters for the new construction section to initialize the meta-model, and achieve cross-section migration and deployment of the model through lightweight adjustment, so as to complete the multi-modal intelligent collaborative suppression of wind-induced vibration during the construction period of the cable-stayed bridge.

[0067] Specifically, the mapping calculation module 701 includes: The first generation unit: It is used to obtain the three-dimensional wind speed vector field data of the construction area by using a lidar array with a spatial resolution of 0.5 m and a time sampling interval of 0.1 s, and real-time monitor the strain field of the main girder joint through a distributed fiber Bragg grating sensor network with a sensitivity of 1 microstrain and a sampling frequency of 200 Hz; use the carrier phase differential technology and the 1588 time protocol to map the three-dimensional wind speed vector field data and the strain field data to a unified three-dimensional space-time grid coordinate system to generate a fusion data set; The judgment unit: It is used to perform spatial difference operations on the strain field, calculate the strain gradient vector of each spatial point of the main girder joint strain field, and extract the Euclidean norm in the strain gradient vector as the gradient amplitude, and judge whether the gradient amplitude exceeds 50 με / m. If it exceeds, it is marked as a potential stress concentration area. If it does not exceed, it is not marked; calculate the local vorticity component in the three-dimensional wind speed vector field data, set a threshold to judge whether the local vorticity component is greater than the threshold. If it exceeds, it is marked as a strong rotating wind field area. If it does not exceed, it is not marked; The second generation unit: It is used to traverse all the space-time grid points, screen out the points that simultaneously include the potential stress concentration area and the strong rotating wind field area, and determine the grid points that meet the conditions as the high-risk points of wind-structure coupling. Integrate the grid points of all the high-risk points of wind-structure coupling to generate a composite risk focus coordinate set, which is recorded as the priority target area for damper dynamic deployment and aerodynamic fin regulation.

[0068] Specifically, the generation module 702 includes: The acquisition unit: It is used to assign an initial pheromone concentration value to all coordinate points in the composite risk focus coordinate set, construct a bionic pheromone concentration field by using the Gaussian kernel density estimation method, and perform dynamic update by adopting a dual-time scale mechanism suitable for the cantilever construction of the cable-stayed bridge, so as to obtain the updated pheromone concentration field, where the dual-time scale mechanism includes short-time scale update and long-time scale reset; The first calculation unit: It is used to set dynamic update rules according to the updated pheromone concentration field and the strain gradient amplitude, and calculate the damper path probability density. The calculation formula is as follows:

[0069] In the formula, For the priority of deploying a damper at the coordinates ( x , y ), is the bionic pheromone concentration field, is the strain gradient amplitude, is the vorticity threshold function, is the threshold set for the cantilever construction of the steel box girder, is the global integral normalization factor, which integrates over the entire bridge deck area; The second calculation unit: used to combine the damper path probability density and the time-varying model of the main girder stiffness, apply the weighted Voronoi tessellation algorithm to generate a set of responsibility partitions, use the MEMS accelerometer array to collect data at a sampling rate of 200 Hz, and calculate the standard deviation of the acceleration response within each partition; if the standard deviation of any partition exceeds 8%, adjust the tessellation weight and re-tessellate until the standard deviation of the acceleration response of all partitions meets the standard, thereby obtaining a set of partitions that meet the standard.

[0070] Specifically, the establishment module 703, which includes: The first deployment unit: used to deploy a micro pneumatic flap array in the form of an equally spaced matrix within the composite risk focus coordinate set according to the risk assessment within the set of responsibility partitions; The third calculation unit: used to calculate the real-time angle of attack of each pneumatic flap at time t using the target intervention frequency, and its calculation formula is as follows:

[0071] In the formula, is the real-time angle of attack of the th pneumatic flap, is the set value of the initial angle of attack, is the target intervention frequency, t is the time, is the proportional control gain, is the strain value measured at the position and time t , is the Heaviside step function, is the strain gradient amplitude, is the threshold of the strain gradient; The construction unit: used to construct an empirical mapping model based on the relationship between the damping force, current, and vibration response and obtain the optimal current sequence through model predictive control, and its calculation formula is as follows:

[0072] In the formula, is the The current control sequence of a magnetorheological damper, where T is the total time range, i.e., the upper limit of the time step. is the trajectory density distribution. is the energy consumption penalty coefficient. is time t of the current control sequence. is the structural vibration acceleration response. is the relationship between the damping force, current, and vibration response.

[0073] Specifically, the adjustment and optimization module 704 includes: The second deployment unit: used to deploy the MEMS accelerometer array at the key positions of the bridge according to the control instruction. The key positions include the main girder, bridge tower, and stay cable structural components, and record the acceleration data of each sensor. The fourth calculation unit: used to analyze the collected acceleration data, calculate the energy decay rate by comparing the energy changes of the system before and after vibration suppression. The calculation formula is as follows:

[0074] In the formula, is the energy decay rate at time t. is to accumulate the energy changes of all monitoring points on the entire bridge structure. is the velocity response before vibration suppression. α is the strain energy weight coefficient. is the strain history data before vibration suppression. is the acceleration response after vibration suppression. is the real-time strain after vibration suppression. is to calculate the energy change from the initial moment to the current moment. is all the tiny time intervals. The adjustment unit: used to adjust the angle of attack of the pneumatic flap based on the energy decay rate to obtain the adjusted angle of attack direction. The calculation formula is as follows:

[0075] In the formula, is the gradient direction sign function. is the th micro pneumatic flap's angle of attack at time t t. is the th micro pneumatic flap's position coordinate on the bridge structure, and t is the current time point. Simulation and Optimization Unit: It is used to perform simulations in the digital twin model by utilizing the collected real-time wind speed vector field data and bridge structure response data to predict the response of the bridge structure over a period of time in the future. Compare the predicted response obtained from the simulation with the actual monitoring data. If there are differences, perform optimization, and combine the optimization results with the adjusted angle of attack direction to evaluate the current control strategy. Adjustment and Coordination Unit: It is used to respond to the energy decay rate and the current control strategy by dynamically adjusting control parameters. The adjustment of control parameters includes iterative solution of model predictive control, optimization of the angle of attack adjustment strategy for pneumatic fins, and current control sequence of magnetorheological dampers, so as to achieve collaborative suppression of wind vibration and adapt to the current wind field conditions and bridge structure response.

[0076] It should be noted that regarding the system in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0077] Embodiment 3:

[0078] Corresponding to the above method embodiment, in this embodiment, a multi-modal intelligent collaborative suppression device for wind vibration during the construction period of a cable-stayed bridge is also provided. The multi-modal intelligent collaborative suppression device for wind vibration during the construction period of a cable-stayed bridge described below can be mutually referred to with the multi-modal intelligent collaborative suppression method for wind vibration during the construction period of a cable-stayed bridge described above.

[0079] Figure 3 It is a block diagram of a multi-modal intelligent collaborative suppression device 800 for wind vibration during the construction period of a cable-stayed bridge shown according to an exemplary embodiment. As Figure 3 shown, the multi-modal intelligent collaborative suppression device 800 for wind vibration during the construction period of a cable-stayed bridge includes: a processor 801 and a memory 802. The multi-modal intelligent collaborative suppression device 800 for wind vibration during the construction period of a cable-stayed bridge further includes one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0080] Among them, the processor 801 is used to control the overall operation of the multi-modal intelligent collaborative vibration suppression device 800 during the construction period of the cable-stayed bridge, so as to complete all or part of the steps in the above multi-modal intelligent collaborative vibration suppression method during the construction period of the cable-stayed bridge. The memory 802 is used to store various types of data to support the operation of the multi-modal intelligent collaborative vibration suppression device 800 during the construction period of the cable-stayed bridge. These data may include, for example, instructions for any application program or method operating on the multi-modal intelligent collaborative vibration suppression device 800 during the construction period of the cable-stayed bridge, as well as application program-related data, such as contact data, sent and received messages, pictures, audio, video, and so on. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 803 may include a screen and an audio component. Among them, the screen can be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone, and the microphone is used to receive external audio signals. The received audio signals can be further stored in the memory 802 or sent through the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above other interface modules can be a keyboard, a mouse or buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the multi-modal intelligent collaborative vibration suppression device 800 during the construction period of the cable-stayed bridge and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them. Therefore, the corresponding communication component 805 may include: a Wi-Fi module, a Bluetooth module or an NFC module.

[0081] In an exemplary embodiment, the multi-modal intelligent collaborative suppression device 800 for wind-induced vibration during the construction period of a cable-stayed bridge can be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components, and is used to execute the above-mentioned multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge.

[0082] In another exemplary embodiment, there is also provided a computer-readable storage medium including program instructions, and when the program instructions are executed by a processor, the steps of the above-mentioned multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 802 including program instructions, and the above-mentioned program instructions can be executed by the processor 801 of the multi-modal intelligent collaborative suppression device 800 for wind-induced vibration during the construction period of a cable-stayed bridge to complete the above-mentioned multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge.

[0083] Embodiment 4:

[0084] Corresponding to the above method embodiment, in this embodiment, there is also provided a readable storage medium. A readable storage medium described below can be correspondingly referred to with a multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge described above.

[0085] A computer program is stored on the readable storage medium, and when the computer program is executed by a processor, the steps of the multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge in the above method embodiment are implemented.

[0086] The readable storage medium can specifically be various readable storage media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program codes.

[0087] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

[0088] As described above, these are only the specific implementation manners of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or replacements, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge, characterized in that: include: The three-dimensional wind speed vector field data of the construction area is collected by using a laser radar array. At the same time, the strain field of the main beam joint is synchronously monitored by a distributed fiber grating sensor. The three-dimensional wind speed vector field data and the strain field are mapped to a unified three-dimensional space-time grid coordinate system through high-precision time synchronization technology. The strain gradient amplitude is obtained based on spatial difference calculation, and combined with the local vorticity component of the wind speed field and the double threshold criterion, a composite risk focus coordinate set is screened and generated, which is recorded as the target area for dynamic deployment of the damper. Based on the composite risk focus coordinate set, the bionic pheromone concentration field is constructed, and the update rules are set. The path probability density distribution of the damper is generated in combination with the strain gradient amplitude. The weighted Voronoi meshing weight is defined according to the main beam stiffness distribution and the path probability density. The responsibility partition set is generated, and the standard deviation of the vortex vibration energy in each partition is verified. If it is not met, the meshing weight is iteratively adjusted until it meets the standard. The local average wind speed is obtained by processing the three-dimensional wind speed vector field data in space and time, and the target intervention frequency is obtained by inverse calculation through the Strouhal number criterion in combination with the characteristic height of the main beam; according to the responsibility partition set, the micro-aerodynamic wing array is deployed in the composite risk focus coordinate set, and based on the target intervention frequency, the real-time angle of attack command is generated, and an empirical mapping model of the damping force-current-vibration response is established, and the optimal current sequence is solved through model predictive control; based on the time slot allocation protocol of the time-sensitive network, the control command is issued, where the control command is to send the angle of attack command and the current spectrum to the edge computing node, thereby controlling the aerodynamic wing and the magnetorheological damper; According to the control instructions, the MEMS accelerometer array monitors the acceleration response data of the bridge deck after vibration suppression, calculates the energy attenuation rate, and adjusts the angle of attack of the aerodynamic wing accordingly. At the same time, the digital twin technology is used to generate predicted responses, dynamically adjust the control parameters, and form an optimized control strategy. An optimization control strategy is adopted to classify and store historical data according to wind field characteristic parameters to construct a multimodal dataset. MAML meta-learning is used to update model parameters online, and matching wind field parameters are loaded for new construction sections to initialize the meta-model. The model is then migrated and deployed across sections through lightweight adjustments, thereby completing the multimodal intelligent collaborative suppression of wind-induced vibration during the construction period of cable-stayed bridges.

2. The multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge according to claim 1 is characterized in that: The three-dimensional wind speed vector field data of the construction area is collected by using a laser radar array, and the strain field of the main beam joint is synchronously monitored by a distributed fiber grating sensor. The three-dimensional wind speed vector field data and the strain field are mapped to a unified three-dimensional space-time grid coordinate system through high-precision time synchronization technology. The strain gradient amplitude is obtained based on spatial difference calculation, and combined with the local vorticity component of the wind speed field and the double threshold criterion, a composite risk focus coordinate set is screened and generated, which is recorded as the target area for dynamic deployment of the damper, including: The three-dimensional wind speed vector field data of the construction area was acquired by using a lidar array with a spatial resolution of 0.5m and a time sampling interval of 0.1 seconds. The strain field of the main beam joint was monitored in real time through a distributed fiber grating sensor network with 1 micro-strain sensitivity and a sampling frequency of 200Hz. The three-dimensional wind speed vector field data and the strain field data were mapped to a unified three-dimensional space-time grid coordinate system using carrier phase difference technology and 1588 time protocol to generate a fused data set. Perform spatial differential operation on the strain field, calculate the strain gradient vector of each spatial point in the main beam joint strain field, and extract the Euclidean norm in the strain gradient vector as the gradient amplitude to determine whether the gradient amplitude exceeds 50με / m. If it exceeds, it is marked as a potential stress concentration area, and if it does not exceed, it is not marked; calculate the local vorticity component in the three-dimensional wind speed vector field data, set a threshold to determine whether the local vorticity component is greater than the threshold, if it exceeds, it is marked as a strong rotating wind field area, and if it does not exceed, it is not marked; All space-time grid points are traversed to screen out points that include both potential stress concentration areas and strong rotating wind field areas, and the grid points that meet the conditions are determined as high-risk points for wind-structure coupling. The grid points of all high-risk points for wind-structure coupling are integrated to generate a composite risk focus coordinate set, which is recorded as the priority target area for dynamic deployment of dampers and regulation of aerodynamic blades.

3. The multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge according to claim 1 is characterized in that: The bionic pheromone concentration field is constructed based on the composite risk focus coordinate set, and the update rules are set. The path probability density distribution of the damper is generated in combination with the strain gradient amplitude. The weighted Voronoi partition weight is defined according to the main beam stiffness distribution and the path probability density. The responsibility partition set is generated, and the standard deviation of the vortex vibration energy in each partition is verified. If it is not satisfied, the partition weight is iteratively adjusted until it meets the standard, including: In the composite risk focus coordinate set, all coordinate points are assigned initial pheromone concentration values, and the bionic pheromone concentration field is constructed using the Gaussian kernel density estimation method. The dual time scale mechanism suitable for the cantilever construction of cable-stayed bridges is used for dynamic update to obtain the updated pheromone concentration field, where the dual time scale mechanism includes short-time scale update and long-time scale reset. According to the updated pheromone concentration field and strain gradient amplitude, the dynamic update rule is set to calculate the damper path probability density, and the calculation formula is as follows: In the formula, For the coordinates ( x , y ) is the priority of deploying dampers at is the bionic pheromone concentration field, is the strain gradient amplitude, is the vorticity threshold function, The threshold value set for cantilever construction of steel box girders, is the global integration normalization factor, which is integrated over the entire bridge deck area; Combining the probability density of damper paths and the time-varying stiffness model of main beams, the weighted Voronoi partitioning algorithm is applied to generate the responsible partition set. The MEMS accelerometer array is used to collect data at a sampling rate of 200 Hz, and the standard deviation of the acceleration response in each partition is calculated. If the standard deviation of any partition exceeds 8%, the partitioning weight is adjusted and re-partitioned until the standard deviation of the acceleration response of all partitions meets the standard, thus obtaining a partition set that meets the standard.

4. The multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge according to claim 1 is characterized in that: According to the responsibility partition set, a micro-aerodynamic wing array is deployed in the composite risk focus coordinate set, a real-time angle of attack command is generated based on the target intervention frequency, and an empirical mapping model of damping force-current-vibration response is established, and the optimal current sequence is solved through model predictive control, which includes: According to the risk assessment in the responsibility partition set, the micro-aerodynamic wing array is deployed in a matrix with equal spacing in the composite risk focus coordinate set; the target intervention frequency is used to calculate the time of each aerodynamic wing t The real-time angle of attack is calculated as follows: In the formula, For the The real-time angle of attack of each aerodynamic blade, is the setting value of the initial angle of attack, is the target intervention frequency, t is the time, is the proportional control gain, For the location and time t The measured strain value, is the Heaviside step function, is the strain gradient amplitude, is the threshold of strain gradient; Based on the relationship between damping force, current and vibration response, an empirical mapping model is constructed, and the optimal current sequence is obtained through model predictive control. The calculation formula is as follows: In the formula, For the The current control sequence of a magnetorheological damper, T is the total time range, that is, the upper limit of the time step, is the trajectory density distribution, is the energy consumption penalty coefficient, For time t The current control sequence, is the structural vibration acceleration response, is the relationship between damping force, current and vibration response.

5. The multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge according to claim 1 is characterized in that: According to the control instructions, the acceleration response data of the bridge deck after vibration suppression is monitored by the MEMS accelerometer array, the energy attenuation rate is calculated, and the angle of attack of the aerodynamic wing is adjusted accordingly. At the same time, the digital twin technology is used to generate a predicted response, dynamically adjust the control parameters, and form an optimized control strategy, which includes: According to the control instructions, MEMS accelerometer arrays are deployed at key locations of the bridge, including the main beam, bridge towers and cable-stayed structural components, to record the acceleration data of each sensor; The collected acceleration data is analyzed, and the energy attenuation rate is calculated by comparing the energy change of the system before and after vibration suppression. The calculation formula is as follows: In the formula, is the energy decay rate at time t, In order to accumulate the energy changes of all monitoring points on the entire bridge structure, is the velocity response before vibration suppression, α is the strain energy weight coefficient, is the strain history data before vibration suppression, is the acceleration response after vibration suppression, is the real-time strain after vibration suppression, To calculate the energy change from the initial moment to the current moment, For all tiny time intervals; Based on the energy attenuation rate, the angle of attack of the aerodynamic wing is adjusted to obtain the adjusted angle of attack direction, which is calculated as follows: In the formula, is the gradient direction sign function, For the A micro-aerodynamic blade in time t The angle of attack, For the The position coordinates of the micro-aerodynamic wing on the bridge structure, t is the current time point; The collected real-time wind speed vector field data and bridge structure response data are used to simulate in the digital twin model to predict the response of the bridge structure in the future. The predicted response obtained by simulation is compared with the actual monitoring data. If there is a difference, optimization is performed and the optimization result is combined with the adjusted angle of attack direction to evaluate the current control strategy. By dynamically adjusting the control parameters in response to the energy attenuation rate and the current control strategy, the adjustment of the control parameters includes iteratively solving the model predictive control, optimizing the aerodynamic blade angle of attack adjustment strategy and the magnetorheological damper current control sequence, so as to achieve coordinated suppression of wind vibration and adapt to the current wind field conditions and bridge structure response.

6. A multi-modal intelligent collaborative suppression system for wind-induced vibration during the construction period of a cable-stayed bridge, based on the multi-modal intelligent collaborative suppression method for wind-induced vibration during the construction period of a cable-stayed bridge according to claim 1, characterized in that: include: Mapping calculation module: It is used to collect three-dimensional wind speed vector field data in the construction area using a laser radar array, and simultaneously monitor the strain field of the main beam joint through a distributed fiber grating sensor. Through high-precision time synchronization technology, the three-dimensional wind speed vector field data and the strain field are mapped to a unified three-dimensional space-time grid coordinate system. The strain gradient amplitude is obtained based on spatial difference calculation, and combined with the local vorticity component of the wind speed field and the dual threshold criterion, a composite risk focus coordinate set is screened and generated, which is recorded as the target area for dynamic deployment of the damper; Generation module: used to construct the bionic pheromone concentration field based on the composite risk focus coordinate set, set the update rules, generate the path probability density distribution of the damper in combination with the strain gradient amplitude, define the weighted Voronoi partition weight according to the main beam stiffness distribution and the path probability density, generate the responsibility partition set, verify the standard deviation of the vortex vibration energy in each partition, and iteratively adjust the partition weight until it meets the standard if it is not met; Establishing module: It is used to obtain the local average wind speed by processing the three-dimensional wind speed vector field data in space and time, and to obtain the target intervention frequency by inverse calculation through the Strouhal number criterion in combination with the characteristic height of the main beam; deploying the micro-aerodynamic wing array in the composite risk focus coordinate set according to the responsibility partition set, generating real-time angle of attack instructions based on the target intervention frequency, and establishing an empirical mapping model of damping force-current-vibration response, and solving the optimal current sequence through model predictive control; issuing control instructions based on the time slot allocation protocol of the time-sensitive network, wherein the control instructions are to issue the angle of attack instructions and the current spectrum to the edge computing node, thereby controlling the aerodynamic wing and the magnetorheological damper; Adjustment and optimization module: It is used to monitor the acceleration response data of the bridge deck after vibration suppression through the MEMS accelerometer array according to the control instructions, calculate the energy attenuation rate, and adjust the angle of attack of the aerodynamic wing accordingly. At the same time, it uses digital twin technology to generate predicted responses, dynamically adjust control parameters, and form an optimized control strategy; Construction module: It is used to adopt an optimization control strategy, classify and store historical data according to wind field characteristic parameters, and construct a multimodal data set; use MAML meta-learning to update model parameters online, load matching wind field parameters for new construction sections to initialize the meta-model, and realize cross-section migration and deployment of the model through lightweight adjustment, thereby completing the multimodal intelligent collaborative suppression of wind vibration during the construction period of the cable-stayed bridge.

7. The multi-modal intelligent coordinated suppression system for wind-induced vibration during the construction period of a cable-stayed bridge according to claim 6 is characterized in that: The mapping calculation module includes: The first generation unit is used to obtain the three-dimensional wind speed vector field data of the construction area with a spatial resolution of 0.5m and a time sampling interval of 0.1 seconds using a laser radar array, and to monitor the strain field of the main beam joint in real time through a distributed fiber grating sensor network with 1 micro-strain sensitivity and a sampling frequency of 200Hz; the three-dimensional wind speed vector field data and the strain field data are mapped to a unified three-dimensional space-time grid coordinate system using carrier phase difference technology and 1588 time protocol to generate a fused data set; Judgment unit: used to perform spatial differential operation on the strain field, calculate the strain gradient vector of each spatial point in the strain field of the main beam joint, and extract the Euclidean norm in the strain gradient vector as the gradient amplitude, and judge whether the gradient amplitude exceeds 50με / m. If it exceeds, it is marked as a potential stress concentration area, and if it does not exceed, it is not marked; calculate the local vorticity component in the three-dimensional wind speed vector field data, set a threshold to judge whether the local vorticity component is greater than the threshold, if it exceeds, it is marked as a strong rotating wind field area, and if it does not exceed, it is not marked; The second generation unit is used to traverse all space-time grid points, screen out points that include both potential stress concentration areas and strong rotating wind field areas, and determine the grid points that meet the conditions as high-risk points for wind-structure coupling. The grid points of all wind-structure coupling high-risk points are integrated to generate a composite risk focus coordinate set, which is recorded as the priority target area for dynamic deployment of dampers and regulation of aerodynamic blades.

8. The multi-modal intelligent coordinated suppression system for wind-induced vibration during the construction period of a cable-stayed bridge according to claim 6 is characterized in that: The generation module includes: Acquisition unit: used to assign initial pheromone concentration values ​​to all coordinate points in the composite risk focus coordinate set, construct the bionic pheromone concentration field using the Gaussian kernel density estimation method, and dynamically update it using a dual-time scale mechanism suitable for the cantilever construction of cable-stayed bridges, thereby obtaining the updated pheromone concentration field, wherein the dual-time scale mechanism includes short-time scale update and long-time scale reset; The first calculation unit is used to set dynamic update rules according to the updated pheromone concentration field and strain gradient amplitude, and calculate the damper path probability density. The calculation formula is as follows: In the formula, For the coordinates ( x , y ) is the priority of deploying dampers at is the bionic pheromone concentration field, is the strain gradient amplitude, is the vorticity threshold function, The threshold value set for cantilever construction of steel box girders, is the global integration normalization factor, which is integrated over the entire bridge deck area; The second calculation unit is used to combine the damper path probability density and the main beam stiffness time-varying model, apply the weighted Voronoi partitioning algorithm to generate the responsibility partition set, use the MEMS accelerometer array to collect data at a sampling rate of 200 Hz, and calculate the standard deviation of the acceleration response in each partition; if the standard deviation of any partition exceeds 8%, adjust the partition weight and re-partition until the acceleration response standard deviation of all partitions meets the standard, thereby obtaining a partition set that meets the standard.

9. The multi-modal intelligent coordinated suppression system for wind-induced vibration during the construction period of a cable-stayed bridge according to claim 6 is characterized in that: The establishment module includes: The first deployment unit is used to deploy the micro-aerodynamic wing array in a matrix form with equal spacing in the composite risk focus coordinate set according to the risk assessment in the responsibility partition set; The third calculation unit is used to calculate the time of each aerodynamic wing using the target intervention frequency. t The real-time angle of attack is calculated as follows: In the formula, For the The real-time angle of attack of each aerodynamic blade, is the setting value of the initial angle of attack, is the target intervention frequency, t is the time, is the proportional control gain, For the location and time t The measured strain value, is the Heaviside step function, is the strain gradient amplitude, is the threshold of strain gradient; Construction unit: used to construct an empirical mapping model based on the relationship between damping force, current and vibration response, and obtain the optimal current sequence through model predictive control. The calculation formula is as follows: In the formula, For the The current control sequence of a magnetorheological damper, T is the total time range, that is, the upper limit of the time step, is the trajectory density distribution, is the energy consumption penalty coefficient, For time t The current control sequence, is the structural vibration acceleration response, is the relationship between damping force, current and vibration response.

10. The multi-modal intelligent coordinated suppression system for wind-induced vibration during the construction period of a cable-stayed bridge according to claim 6, characterized in that: The adjustment and optimization module includes: The second deployment unit is used to deploy MEMS accelerometer arrays at key locations of the bridge according to control instructions, where the key locations include the main beam, bridge towers and cable-stayed structural components, and record the acceleration data of each sensor; The fourth calculation unit is used to analyze the collected acceleration data and calculate the energy attenuation rate by comparing the energy change of the system before and after vibration suppression. The calculation formula is as follows: In the formula, is the energy decay rate at time t, In order to accumulate the energy changes of all monitoring points on the entire bridge structure, is the velocity response before vibration suppression, α is the strain energy weight coefficient, is the strain history data before vibration suppression, is the acceleration response after vibration suppression, is the real-time strain after vibration suppression, To calculate the energy change from the initial moment to the current moment, For all tiny time intervals; Adjustment unit: used to adjust the angle of attack of the aerodynamic wing based on the energy attenuation rate to obtain the adjusted angle of attack direction. The calculation formula is as follows: In the formula, is the gradient direction sign function, For the A micro-aerodynamic blade in time t The angle of attack, For the The position coordinates of the micro-aerodynamic wing on the bridge structure, t is the current time point; Simulation optimization unit: used to use the collected real-time wind speed vector field data and bridge structure response data to simulate in the digital twin model to predict the response of the bridge structure in the future, compare the predicted response obtained by simulation with the actual monitoring data, and optimize if there is a difference. The optimization result is combined with the adjusted angle of attack direction to evaluate the current control strategy; Adjustment coordination unit: used to dynamically adjust control parameters in response to the energy attenuation rate and the current control strategy. The adjustment of control parameters includes iterative solution of model predictive control, optimization of aerodynamic blade angle of attack adjustment strategy and magnetorheological damper current control sequence, so as to achieve coordinated suppression of wind vibration and adapt to the current wind field conditions and bridge structure response.

Images