Real-time monitoring and regulation method and system for installation process of coupled beam damper

Abstract

The application discloses a real-time monitoring and regulation method and system for a coupling beam damper installation process, and belongs to the field of building structure construction installation monitoring and intelligent regulation. In order to solve the problems that the six-degree-of-freedom deviation and the fastening force pre-tightening force are difficult to be continuously and quantitatively monitored and closed-loop deviation is corrected during installation of the coupling beam damper, the application establishes an installation digital reference model, collects position and posture and pre-tightening force data in real time, constructs an installation state factor graph and performs incremental optimization, generates a constraint regulation instruction in combination with coupling sensitivity self-correction and confidence weight, and realizes closed-loop control of installation deviation, pre-tightening force and dispersion.

Description

Technical Field

[0001] This invention relates to the field of monitoring and intelligent control of building structure construction and installation, and in particular to a real-time monitoring and control method and system for the installation process of a coupling beam damper. Background Technology

[0002] Coupling beam dampers are typically installed in coupling beams or related connection nodes of building structures to improve structural energy dissipation and seismic performance. Their installation quality depends not only on the geometrical accuracy during hoisting, alignment, and fixing, but also on the fit of the connecting parts at both ends, the correspondence of bolt holes, and the preload of high-strength bolts or anchors. Currently, on construction sites, manual measurement, levels, or local sensor readings are mostly used for phased inspections, and installation adjustments primarily rely on construction experience and judgment of individual deviations.

[0003] In the above-mentioned methods, the axial, lateral, and vertical displacement deviations, as well as multiple angular deviations, between the damper body, the end connectors, and the connecting beam nodes are difficult to quantify continuously under the same coordinate reference. Furthermore, there is a lack of unified integration between preload data and geometric deviation data. Construction personnel typically perform jacking, pulling back, shim adjustment, or graded tightening after discovering local deviations. However, the coupled response of these control actions to changes in deviation and preload is affected by on-site stiffness, clearance, friction, and tightening sequence, making accurate prediction difficult based solely on experience.

[0004] Therefore, the existing installation process is prone to problems such as deviations and inconsistencies in preload that are difficult to identify in a timely manner. This may lead to risks such as hard-top installation, unilateral stress, slippage and loosening, and abnormal initial internal forces after installation, affecting the installation acceptance of the connecting beam damper and the stability of its subsequent service status.

[0005] Therefore, there is a need for a real-time monitoring and control method and system for the installation process of coupling beam dampers that can overcome the shortcomings of the existing technologies. Summary of the Invention

[0006] One objective of this invention is to propose a real-time monitoring and control method and system for the installation process of a connecting beam damper. Addressing the problem in existing technologies where it is difficult to continuously and quantitatively monitor and correct six-degree-of-freedom installation deviations and fastener preload during the hoisting, alignment, and final tightening of the connecting beam damper, this invention proposes a technical solution that involves establishing an installation digital reference model, collecting pose and preload data, constructing an installation state factor diagram, recursively correcting the coupling sensitivity matrix, and solving the constraint control model. This invention achieves the technical effect of enabling installation deviations, preload values, and preload dispersion to be monitored, predicted, and controlled within the same closed loop.

[0007] This invention provides a real-time monitoring and control method for the installation process of a connecting beam damper, comprising: S1, establishing an installation digital reference model based on the damper's design axis, the mounting surfaces of the connectors at both ends, bolt hole positions, elevation reference, installation tolerance, preload design value, and fastening sequence constraints, generating an installation reference coordinate system, a geometric constraint set, and a preload constraint set; S2, collecting the pose data and preload data of the damper body, connectors at both ends, connecting beam connection nodes, and each fastener, and transforming them to the installation reference coordinate system to obtain unified installation observation data; S3, constructing an installation state factor diagram using the unified installation observation data, geometric constraint set, and preload constraint set, and incrementally optimizing to obtain a six-degree-of-freedom installation deviation vector and fastener... S4. Based on the executed control amount and the deviation change and preload change obtained from the retest, the coupling sensitivity matrix is ​​recursively corrected, and confidence weights are generated based on the sensor residuals. S5. Based on the installation stage, the six-degree-of-freedom installation deviation vector, the fastener preload state vector, confidence weights, the coupling sensitivity matrix, the geometric constraint set, and the preload constraint set, the constraint control model is solved, and control instructions including the pushing direction, pulling direction, step amount, gasket thickness combination, fastening sequence, graded loading parameters, and retest points are generated and executed. The retest data is fed back to the installation state factor diagram until the six-degree-of-freedom installation deviation vector and the fastener preload state vector satisfy the geometric constraint set and the preload constraint set.

[0008] Optionally, S1 includes:

[0009] Establish an installation reference coordinate system using the construction measurement benchmark of the coupling beam as the coordinate reference;

[0010] The damper design axis and the mounting surfaces of the connectors at both ends are converted into axial constraints and end face plane constraints, respectively.

[0011] The bolt hole positions, elevation datum, and installation tolerances are converted into hole position coincidence constraints, elevation constraints, and six-degree-of-freedom tolerance boundaries.

[0012] Convert the design value of preload, the allowable deviation of preload, and the limit of preload dispersion into a set of preload constraints;

[0013] The fastening sequence constraints are converted into a stage sequence table for fastener groups to obtain the installation digital reference model.

[0014] Optionally, S2 includes:

[0015] Raw measurement data are obtained by using pose targets set on the damper body and the connecting parts at both ends, reference targets set on the connecting beam nodes, and preload acquisition units for fasteners.

[0016] The raw measurement data is timestamped and the equipment coordinates are calibrated.

[0017] The pose data is converted into the pose of the damper body, the pose of the two end connectors and the pose of the connecting beam node by using the external parameter transformation between the installation reference coordinate system and the coordinate system of each acquisition device.

[0018] The preload data is mapped to preload observation values ​​according to the fastener number, forming unified installation observation data.

[0019] Optionally, S3 includes:

[0020] The damper body pose, the poses of the two end connectors, the poses of the connecting beam node, and the preload of each fastener are used as variable nodes.

[0021] The pose observation, end face fit, axis coaxiality, hole position correspondence, elevation, fastening sequence and preload design constraints are respectively treated as factor nodes;

[0022] Set residual weights for factor nodes according to confidence weights or preset measurement weights;

[0023] Using the previous round of optimization results as a priori, the installation state factor diagram is solved incrementally, and the output consists of a six-degree-of-freedom installation deviation vector composed of axial displacement deviation, lateral displacement deviation, vertical displacement deviation, rotation angle deviation around the axis, rotation angle deviation around the lateral angle deviation and rotation angle deviation around the vertical angle deviation, as well as a fastener preload state vector composed of the estimated values ​​of each fastener preload.

[0024] Optionally, S4 includes:

[0025] The executed push, pullback, temporary support lifting and lowering, shim thickness combination, and tightening increment are encoded into the control input increment vector;

[0026] The difference between the six-degree-of-freedom installation deviation vectors before and after the retest and the difference between the fastener preload state vector are encoded as the installation response increment vector.

[0027] The update samples are constructed using the incremental vector of the control input and the incremental vector of the installation response. The coupling sensitivity matrix is ​​updated using recursive least squares with a forgetting factor, and the update results are constrained by element limiting and sign consistency verification.

[0028] The updated coupling sensitivity matrix is ​​input into the constraint control model in step S5 to predict the impact of the next round of control commands on the six-degree-of-freedom installation deviation vector and the fastener preload state vector.

[0029] Furthermore, the confidence weights are generated in the following way: the pose observation residual, the preload observation residual, the change in the same measurement point at adjacent times, and the observation difference of different sensors at the same measurement point are normalized to obtain the measurement residual component, the time consistency component, and the sensor consistency component.

[0030] The basic confidence weights of each observation factor are generated based on the preset noise threshold and the residual component mapping table.

[0031] When the basic confidence weight of any observation factor is lower than the retest trigger threshold, the observation factor is marked as a factor to be retested and its residual weight is restricted when entering S3.

[0032] If the observed factor is still below the retest trigger threshold after retesting, it will be removed from the target item of the current constraint control model. The confidence weight is used to adjust the weight of the target item.

[0033] Optionally, S5 includes:

[0034] A constrained control model is established with the six-degree-of-freedom installation deviation weighted quantity, preload deviation weighted quantity, preload dispersion, and control input change quantity as objective terms;

[0035] The geometric constraint set, the preload constraint set, the stroke boundary of the jacking mechanism, the bearing boundary of the temporary support, the discrete thickness set of the gaskets, and the fastening sequence table are used as constraint conditions.

[0036] The six-degree-of-freedom installation deviation vector and fastener preload state vector corresponding to the candidate control command are predicted based on the coupling sensitivity matrix, and the weights of the target terms are adjusted according to the confidence weights.

[0037] Select the control command whose target value satisfies the convergence criterion from the candidate control commands that meet the constraints, and output the pushing direction, pulling direction, step amount, shim thickness combination, fastening sequence, graded loading parameters and retest points;

[0038] Furthermore, the execution and feedback of control commands include: breaking down control commands into hoisting and positioning control, alignment control, initial tightening control, and final tightening control according to the installation stage;

[0039] After each action in the jacking, pulling back, temporary support raising and lowering, shim addition and reduction and graded tightening is completed, retest data is obtained according to the retest points;

[0040] The retested data will be used as a new observation factor and merged with the previous round of installation status factor diagram;

[0041] When each component of the six-degree-of-freedom installation deviation vector is within the corresponding tolerance boundary, the estimated value of each fastener preload is within the corresponding preload range, and the preload dispersion does not exceed the dispersion limit, an installation closed-loop completion marker is generated.

[0042] If the conditions are not met, the newly added observation factors will be returned to S3 and the next round of optimization and regulation will continue;

[0043] Furthermore, the fastening sequence and graded loading parameters are determined as follows: the fasteners are divided into symmetrical fastening groups and retest fastening groups based on the fastening sequence table in the installation digital reference model;

[0044] Calculate the intra-group and inter-group preload deviations of each symmetrical fastening group based on the fastener preload state vector;

[0045] The loading increment, holding time, and retesting points for each tightening stage are determined based on the preload design value, intra-group preload deviation, inter-group preload deviation, and coupling sensitivity matrix.

[0046] When the predicted loading increment causes any six-degree-of-freedom installation deviation component to exceed the tolerance boundary, the loading increment of the corresponding fastening stage is reduced and the correction control input is added to the next round of constraint control model.

[0047] On the other hand, the present invention also provides a real-time monitoring and control system for the installation process of a coupling beam damper, comprising:

[0048] The module establishes a baseline model to create a digital baseline model for installation based on the damper's design axis, the mounting surfaces of the connectors at both ends, bolt hole positions, elevation baseline, installation tolerance, preload design value, and fastening sequence constraints. It also generates an installation baseline coordinate system, a geometric constraint set, and a preload constraint set. The installation data acquisition module collects the pose data and preload data of the damper body, connectors at both ends, connecting beam nodes, and each fastener, and outputs unified installation observation data. The factor graph optimization module constructs an installation state factor graph and incrementally optimizes it to obtain a six-degree-of-freedom installation deviation vector and a fastener preload state vector. The self-calibration and confidence assessment module recursively corrects the coupling sensitivity matrix and generates confidence weights based on sensor residuals. The constraint control closed-loop module generates and executes control commands based on the installation stage, the six-degree-of-freedom installation deviation vector, the fastener preload state vector, confidence weights, the coupling sensitivity matrix, the geometric constraint set, and the preload constraint set, feeding back the re-measured data to the installation state factor graph.

[0049] The beneficial effects of this invention are:

[0050] 1. By establishing an installation digital reference model that includes the damper design axis, end face, hole position, elevation, installation tolerance, preload design value, and fastening sequence constraints, and by uniformly transforming the collected data to the installation reference coordinate system, the geometric relationship between the damper body, the end connectors, and the connecting beam connection nodes can be formed into a unified calculable object, thereby reducing the risk of deviations and omissions caused by manual and decentralized measurements.

[0051] 2. By constructing an installation state factor diagram using unified installation observation data, geometric constraint set, and preload constraint set, and incrementally optimizing to obtain a six-degree-of-freedom installation deviation vector and fastener preload state vector, the geometric alignment state and fastening state can be reflected simultaneously, enabling deviation exceeding limits, insufficient preload, or abnormal preload dispersion to be detected in a timely manner during installation.

[0052] 3. By recursively correcting the coupling sensitivity matrix by matching the executed control quantity with the retest response, and generating confidence weights by combining the sensor residuals, the next round of constraint control commands can be automatically updated according to the actual stiffness, clearance, friction and fastening sequence on site, reducing the risks of hard ceiling installation, unilateral stress, slippage and loosening and abnormal initial internal forces after installation. Attached Figure Description

[0053] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0054] Fig. 1 A flowchart of a real-time monitoring and control method for the installation process of a coupling beam damper;

[0055] Fig. 2 This is a flowchart of step S4 of the present invention, which recursively corrects the coupling sensitivity and generates confidence weights. Detailed Implementation

[0056] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0057] refer to Figs. 1-2A real-time monitoring and control method for the installation process of a connecting beam damper includes: S1, establishing an installation digital reference model based on the damper's design axis, the mounting surfaces of the connectors at both ends, bolt hole positions, elevation reference, installation tolerance, preload design value, and tightening sequence constraints, generating an installation reference coordinate system, a geometric constraint set, and a preload constraint set; S2, collecting the pose data and preload data of the damper body, connectors at both ends, connecting beam connection nodes, and each fastener, transforming them to the installation reference coordinate system to obtain unified installation observation data; S3, constructing an installation state factor diagram using the unified installation observation data, geometric constraint set, and preload constraint set, and incrementally optimizing to obtain a six-degree-of-freedom installation deviation vector and fastener preload. Force state vector; S4, based on the executed control amount and the deviation change and preload change obtained from the retest, recursively correct the coupling sensitivity matrix, and generate confidence weights based on the sensor residuals; S5, based on the installation stage, the six-degree-of-freedom installation deviation vector, the fastener preload state vector, the confidence weights, the coupling sensitivity matrix, the geometric constraint set, and the preload constraint set, solve the constraint control model, generate and execute control commands including the jacking direction, pullback direction, step amount, gasket thickness combination, fastening sequence, graded loading parameters, and retest points, and feed back the retest data to the installation state factor diagram until the six-degree-of-freedom installation deviation vector and the fastener preload state vector satisfy the geometric constraint set and the preload constraint set.

[0058] In this specific embodiment, S1 includes:

[0059] Establish an installation digital reference model based on the damper design axis, mounting surfaces of the two end connectors, bolt hole positions, elevation reference, installation tolerance, preload design value, and fastening sequence constraints given in the construction drawings, BIM model, or CAD model, and store the installation digital reference model as a data object that can be directly called in subsequent steps;

[0060] The installation of the digital benchmark model is denoted as Its structure is as follows, where each character has a unique meaning within this paragraph and remains consistent with subsequent steps:

[0061] ;

[0062] This indicates the installation of a digital reference model. Indicates the installation reference coordinate system. Represents a set of geometric constraints. Represents the preload constraint set. This represents the stage sequence list for the fastener group;

[0063] Install the reference coordinate system The coordinate system is constructed using the construction survey datum of the coupling beam as a reference. This datum is determined by at least three non-collinear control points provided by the field survey control network. The control point located near the coupling beam connection node is used as the origin of the coordinate system and recorded as the installation datum origin. Simultaneously, the direction of the construction survey datum line of the coupling beam is defined as the horizontal principal axis of the installation datum coordinate system, and the vertical direction of the construction elevation datum is defined as the vertical principal axis of the installation datum coordinate system. The direction of the third axis is determined according to the right-hand rule. This ensures that all geometric elements of the damper body, the end connectors, and the coupling beam connection node are aligned. The following are expressed using the same units of length and angle, and The dimensions and orientation are consistent with the calibration results of the field measuring equipment;

[0064] Geometric constraint set The constraints consist of axis constraints, end face plane constraints, hole coincidence constraints, elevation constraints, and six-degree-of-freedom tolerance boundaries. The axis constraints are derived from the damper design axis, which is represented in the digital model as a two-point spatial straight line. The following record is "Design Axis Element". At the same time, the coaxiality allowable value is written for the axis constraint and the participating object is specified as the target axis of the damper body axis and the connection node of the connecting beam. This enables subsequent steps to calculate the comprehensive deviation in the axial, lateral and vertical directions with "coaxiality" as the target.

[0065] The end face planar constraint is derived from the mounting surfaces of the two end connectors. These mounting surfaces are represented as planar elements in the digital model and... The following record is "End Face Plane Element". At the same time, parameters for the allowable angle of consistency of the end face normal direction and the allowable gap of end face fit are written into the constraint of each end face plane. The participating objects are specified as the mounting surface corresponding to the end face of the connector and the connection node of the connecting beam, so that the subsequent steps can calculate the rotation deviation and end face gap deviation with "fit" as the target.

[0066] The hole coincidence constraint is obtained by converting the bolt hole positions. The bolt hole positions are expressed in the digital model as the coordinates of the hole center and the direction of the hole axis. The following is recorded as "hole position element". At the same time, a hole pair mapping table is established based on the one-to-one correspondence between the design hole number and the construction hole number. The parameters of the allowable offset of the hole center and the allowable deflection angle of the hole axis are written for each hole pair, so that the subsequent steps can constrain the in-plane misalignment and hole axis inconsistency during the installation process with "hole position correspondence" as the target.

[0067] Elevation constraints are derived from elevation datums, where the elevation datum is located at... The target elevation value of the specified measuring point or the specified mounting surface feature point is recorded below, and the allowable elevation deviation parameter is written, so that subsequent steps can constrain the vertical displacement deviation with "elevation met" as the target.

[0068] The six-degree-of-freedom tolerance boundaries are derived from installation tolerances. Installation tolerances are given in process or design documents as axial displacement tolerance, lateral displacement tolerance, vertical displacement tolerance, tolerance around the axis, tolerance around the lateral angle, and tolerance around the vertical angle. The above six tolerance boundaries are written in the form of upper and lower limits using the same symbol. It is then bound to the six-degree-of-freedom deviation object of the damper body relative to the connecting node of the connecting beam, so that subsequent steps can directly determine whether it exceeds the limit and use it as the source of hard constraints for the constraint control model.

[0069] Preload constraint set The preload design value, allowable preload deviation, and preload dispersion limit are converted from the preload design value, allowable preload deviation, and preload dispersion limit. A unique number is assigned to each fastener based on the fastener list, and its preload design value is linked to it. Simultaneously, the allowable preload deviation is written as the upper and lower bounds of the allowable preload range for that fastener, and the preload dispersion limit is written as the intra-group consistency control threshold for symmetrical fastening groups within the same fastening stage. The calculation rules for preload dispersion are as follows: The intermediate curing is defined as "the ratio of the difference between the maximum and minimum preload values ​​within the same symmetrical fastening group to the design value of the preload for that group", and this ratio must not exceed the dispersion limit, so that subsequent steps can simultaneously control the absolute value of the preload and the consistency within the group.

[0070] Fastener group phase sequence table Derived from the fastening sequence constraint, based on the construction process documents, all fasteners are divided into symmetrical fastening groups according to their symmetrical relationship, and each group is assigned a group number. Simultaneously, the installation process is divided into several fastening stages, and for each stage, the allowed set of fastening groups, the order of fastening within each group, and the sequential relationship between stages are written. This ensures that subsequent steps, when solving for control commands, can limit the fastening actions to... Given the phase order and the intra-group order range, and with The pre-tightening constraint enables consistent referencing and consistent judgment.

[0071] In this specific embodiment, S2 includes:

[0072] Based on the installation reference coordinate system Complete the installation data collection and unified observation data generation in conjunction with the fastener numbering system;

[0073] The damper body and the connecting parts at both ends are fixedly installed with coded pose targets, and the connecting beam connection nodes are fixedly installed with reference targets. The coded pose targets and the reference targets are both rigid targets with unique identification codes, and each target contains at least three non-collinear feature points to ensure that the pose can be calculated. The three-dimensional coordinates of the feature points in the target's own coordinate system are given by the factory calibration and written into the target library of the acquisition software.

[0074] The pose data acquisition equipment uses industrial vision measurement equipment and establishes an equipment coordinate system. The collected data includes the unique identifier of each target in each frame, pixel observations of each feature point, and the original target pose measurement results calculated internally by the device. The original target pose measurement results are represented as follows: ,in for Homogeneous transformation matrix including rotation matrix With translation vector Indicates the object being tested and Indicates the damper body, and These represent the connectors at both ends. Indicates the connection node of the coupling beam. Indicates the first The next sampling time;

[0075] The fastener preload acquisition unit uses a bolt axial force sensing washer with a strain measurement bridge and establishes a fastener preload data channel through an acquisition gateway. The acquired data includes the fastener number. The original electrical signal and the preload observation value converted from the calibration curve. and sampling time ,in The unit is N, fastener number With preload constraint set Using the same numbering rule and installed digital reference models Provide a mapping table to ensure that the observations are consistent with the constraints;

[0076] To achieve timestamp alignment of multi-source data, both the pose data acquisition device and the preload acquisition gateway are connected to the same field time reference service and use a unified clock broadcast by the controller as the timestamp source. Simultaneously, in each measurement cycle, the controller sends a synchronization trigger signal and records the trigger edge time as an alignment anchor point. After aligning with the anchor point, the acquisition software interpolates or holds the pose and preload data on the same aligned time sequence to generate a unified sampling time. Data packets, thereby ensuring the same The pose observation and preload observation correspond to the same installation state;

[0077] To realize the equipment coordinate system To install the reference coordinate system External parameter calibration involves establishing a reference rigid body at the connection node of the coupling beam using a benchmark target. Given the known three-dimensional coordinates of the characteristic points of the reference rigid body, the acquisition software acquires multiple frames of observations of the reference target under calibration conditions and uses least-squares rigid body registration to solve for the transformation of the equipment's external parameters. ,in Indicates from the device coordinate system To install the reference coordinate system The homogeneous transformation matrix remains unchanged throughout the installation process and is written into the system configuration;

[0078] After completing timestamp alignment and extrinsic parameter calibration, any object At the alignment time The pose is transformed from the device coordinate system to the installation reference coordinate system, and the transformation relationship is as follows:

[0079] ;

[0080] In the formula Representation Object Install the reference coordinate system The pose homogeneous transformation matrix is ​​given below. This represents the homogeneous transformation matrix of the device's extrinsic parameters. Representation Object Relative to the device coordinate system The original pose homogeneous transformation matrix, Indicates the damper body Connectors at both ends and Or connecting beam nodes Indicates the sampling time after alignment. Let be a homogeneous transformation matrix and its rotation part be Used to characterize the attitude relationship around each axis of the mounting reference coordinate system, its translation component is Used to characterize the three-dimensional position in the mounting reference coordinate system;

[0081] The result , According to the fastener number Organizational preload observation values Encapsulated as unified installation observation data ,in Includes object identifiers, alignment times, sets of homogeneous pose transformation matrices, and sets of preload observations, and adopts the same digital reference model as the installation. Consistent object identification and fastener numbering ensure that subsequent S3 can directly construct the installation state factor diagram and perform incremental optimization.

[0082] In this specific embodiment, S3 includes:

[0083] Based on the installation digital benchmark model and unified installation of observation data Construct an installation state factor diagram and perform incremental optimization to obtain a six-degree-of-freedom installation deviation vector and a fastener preload state vector;

[0084] The installation state factor diagram is denoted as... The variable nodes are determined by the pose of the damper body. Position of the connectors at both ends and Position of connecting beam nodes and the preload condition of all fasteners. Composition, and fastener number and Preload constraint set in Using the same numbering rule and by Provides a collection of fasteners Ensure each Each has a unique constraint object;

[0085] The factor node consists of pose observation factor, end face fit factor, axis coaxiality factor, hole position correspondence factor, elevation factor, fastening sequence factor, and preload design factor. The pose observation factor is composed of... In and The residual is directly generated and the "relative pose error between the estimated pose and the observed pose" is used as the residual. The residual is obtained by performing Lie algebra mapping on the relative pose to obtain a six-dimensional residual vector, which simultaneously represents the three-dimensional translation error and the three-dimensional rotation error.

[0086] End face bonding factor is The end face plane elements are generated and the "normal distance between the end face of the connector and the corresponding mounting surface of the connecting beam" and the "angle between the normals of the two planes" are used as residuals to simultaneously constrain the fitting gap and attitude deflection.

[0087] The coaxiality factor of the axis is determined by The design axis elements are generated and the "shortest distance between the damper body axis and the target axis of the connecting beam node" and the "angle between the axis directions" are used as residuals to constrain axis offset and prevent the axis from being collinear.

[0088] The pore position correspondence factor is determined by The hole mapping table is generated and the "relative displacement of hole center" and "angle of hole axis direction" are used as residuals to constrain hole misalignment and hole axis deflection.

[0089] Elevation factor is The elevation benchmark is generated and the "difference between the estimated elevation of the specified feature point or the specified mounting surface feature point and the target elevation" is used as the residual to constrain vertical displacement.

[0090] Tightening sequence factor is Stage sequence table of fastener groups Generate and be used to define this moment. Allowing entry into the optimized and updated set of fasteners and the fastening stage that allows the application of preload constraints, thereby enabling fasteners in preload states that have not yet reached the stage. Only the prior knowledge factors that are not involved in the current round of preload design factor constraints are retained;

[0091] Preload design factor is Generate and use " and The difference between the design values ​​of the preload force of the corresponding fasteners and " and The difference between the observed values ​​of the preload of the corresponding fasteners is used as the residual to achieve consistency between the design objectives and the sensor observations;

[0092] Set residual weights for each factor node ,in The measurement weights are given by confidence weights or preset measurement weights. The preset measurement weights are used when the system runs for the first time and are overwritten and updated by confidence weights in subsequent rounds. The preset measurement weights are configured with fixed weights for three types of factors: "pose observation factor, geometric constraint factor, and preload design factor" and written into the system configuration file to ensure the consistency of weights among factors of the same type.

[0093] Incremental optimization of the installation state factor diagram is implemented by "using the optimization result of the previous time step as a priori and superimposing the newly added observation factors and newly added constraint factors of the current time step". Specifically, the optimal estimate of the previous time step is used as the priori. Added as a priori factor Then The observed factors obtained by analysis and Add the constraint factors obtained from the analysis The optimal state estimate at this moment is obtained by solving the following equation. :

[0094] ;

[0095] In the formula, Indicates alignment time Estimation of the optimal installation state Represents the set of installation state variables to be estimated and includes With all , This represents the independent variable that minimizes the objective function. Indicates time Enter the set of factor indices to be solved. Indicates the first Factor index, Indicates the first The residual weights of each factor, Indicates the first The residual function of each factor is determined by the geometric relationship between the observed factor or constraint factor and the preload relationship. This indicates that observation data will be installed uniformly. This indicates the installation of a digital reference model. The 2-norm is used to convert the residual vector into a scalar cost. Indicates the first One aligned sampling time;

[0096] The incremental solution is achieved by sparse linearization and iterative update. That is, for the current linearization point, all residual functions are linearized to form a sparse normal equation and an update solution is performed. Then, the updated state is used as the new linearization point for iteration until the preset upper limit of the number of iterations or the threshold of the decrease in the objective function is met. In addition, the sparse structure of the previous time step is reused in each solution to reduce redundant calculations.

[0097] In obtaining Afterwards, based on The design axis elements determine the unit axial direction vector and are based on Determine the vertical unit direction vector, and obtain the horizontal unit direction vector by cross product of the two. Then, combine the estimated relative pose of the damper body with the connecting node of the connecting beam. The target's relative pose is compared to calculate and output a six-degree-of-freedom installation deviation vector consisting of axial displacement deviation, lateral displacement deviation, vertical displacement deviation, rotational deviation around the axis, rotational deviation around the lateral angle, and rotational deviation around the vertical angle. At the same time, the preload of all fasteners is adjusted. Assemble according to fastener number sequence and output fastener preload state vector. .

[0098] In this specific embodiment, S4 includes:

[0099] Installation deviation vector with six degrees of freedom With fastener preload state vector For the installation status results, the control input is the record of the jacking, pulling back, temporary support lifting, shim adjustment and tightening loading actions performed in the previous round of S5, and the retest data is immediately organized and entered into S3 to form a new system after each action. and This allows for the construction of time-aligned samples used to couple sensitivity self-calibration with confidence weight generation.

[0100] The control input increment vector is denoted as Its component is determined by the pushing amount. , pullback amount Temporary support lifting capacity Incremental vector of gasket thickness combination and fastening increment vector It is assembled in a fixed order, among which The unit is Each component corresponds to the installation of a digital reference model. The predefined gasket placement position is used, and the value is the total thickness of the gaskets added or removed in this round at that position, expressed in mm. Each component corresponds to a fastener group's stage sequence table One of the fastening groups is used, and the value is taken as the target preload increment applied to the fastening group in this round, expressed in N;

[0101] The installation response increment vector is denoted as Its variation is determined by the six-degree-of-freedom installation deviation vector. Change in the state vector of fastener preload It is assembled in a fixed order, among which The displacement component is in mm and the rotation component is in mm. The unit is N and the fastener numbering order is the same as the preload constraint set. Consistent;

[0102] The coupling sensitivity matrix is ​​denoted as Its physical meaning is to adjust the input increment near the current installation state. Installation response increment The linear mapping approximation relationship is used to predict the impact of control instructions in S5. S4 employs a recursive least squares method with a forgetting factor. Perform recursive updates and take the forgetting factor. The initial coupling sensitivity matrix is ​​taken as And set the initial covariance matrix. ,in To and Zero matrices of the same dimension To and Identity matrices of the same dimension;

[0103] Recursive updates are performed as follows:

[0104] ,

[0105] ,

[0106] ;

[0107] In the formula, Represents the recursive gain vector. Represents the recursive covariance matrix. This indicates the transpose operation. This indicates that the input increment vector has been adjusted in this round. This represents the incremental vector of the installation response obtained in this round of retesting. The forgetting factor is used to reduce the impact of historical samples on the current update. Indicates the first Alignment time after "execution of control actions and completion of retest feedback";

[0108] To suppress sensitivity divergence caused by abnormal samples, after obtaining Subsequently, element-wise limiting is performed. The upper and lower limits of the limiting are determined by the ratio of "the maximum allowable change of the response component" to "the minimum executable step size of the control input channel" and are fixed in the system configuration. The maximum allowable change of the response component is determined by the geometric constraint set. Six-DOF tolerance boundary and preload constraint set The allowable range of preload force is determined by the joint determination of the preload force range, and the minimum executable step of the control input channel is determined by the minimum step of the jacking mechanism, the minimum step of the pull-back mechanism, the minimum step of the temporary support lifting, the minimum thickness difference of the discrete thickness set of the gasket, and the minimum graded increment of the fastening load.

[0109] Simultaneously, a symbol consistency check is performed to ensure that the update direction meets the on-site physical constraints. The reference symbol matrix for the symbol consistency check is obtained by applying a single-channel positive minimum step to each control input channel and retesting it during the system initialization phase. Confirmed, subsequent updates will be made to... Elements whose signs are opposite to the reference sign and whose magnitudes exceed a preset allowable inverse ratio threshold are backed up, i.e., the element is restored to its original value. The corresponding element is amplified by a fixed proportion in the covariance increment of this element in this round to reduce its impact on subsequent predictions;

[0110] In terms of confidence weight generation, the residuals of each observed factor and constraint factor in S3 are used as inputs, and the output is the confidence weight corresponding to factor one. This is used to simultaneously control the residual weights and the target term weights in S5. The residual vector of each observation factor is denoted as... And estimated by S3 in the optimal installation state The calculation shows that S4 is the pair. The measurement residual component is obtained by normalizing the noise standard deviation corresponding to the factor, and then divided into three weighting levels based on the magnitude of the normalized value. The noise standard deviation is 0.2, where the noise standard deviation is written into the system configuration by the sensor's factory calibration report and field calibration results, and the unit of the translation noise standard deviation of the pose observation factor is mm, the unit of the rotation noise standard deviation is rad, and the unit of the preload observation factor is... ;

[0111] Normalizing the observed changes at adjacent times at the same measuring point yields the time consistency component, where the pose measuring point is represented by... The change in downward translation and the change in equivalent rotation angle are used to determine the value, and the portion exceeding the "allowable change threshold for each measurement cycle" is weighted down. The preload measurement point is determined by the ratio of the difference between two adjacent preload observations to the design value of the fastener's preload, and the portion exceeding this value is weighted down. Partial weight reduction;

[0112] The sensor consistency component is obtained by normalizing the observation differences of different sensors at the same measurement point. The pose measurement point is output by the master pose data acquisition device and the verification pose data acquisition device at the same alignment time. The difference was calculated, and the preload measuring point was determined by the preload acquisition unit. The difference between the calculated value and the converted value of the same fastener loading result recorded by the fastening tool is obtained;

[0113] The basic confidence weight of this factor is obtained by multiplying the measurement residual component, the time consistency component, and the sensor consistency component by a fixed weight. And set a retest trigger threshold. ,when When a factor is marked as a factor to be retested and its residual weight in the next round S3 is fixed at 0.2 times the preset measurement weight of the factor, its impact on the incremental optimization of the factor map is limited. If the basic confidence weight of the factor is still less than [a certain value] after retesting, [further action is required]. This factor is marked as a target elimination factor, and its corresponding target item weight is reset to zero when constructing the target item in the next round of S5, thereby eliminating it from the target item of the current round of constraint control model. At the same time, the updated coupling sensitivity matrix is ​​updated. With all confidence weights Output.

[0114] In this specific embodiment, S5 includes:

[0115] At the alignment time Receive six-degree-of-freedom installation deviation vector With fastener preload state vector Receiver coupling sensitivity matrix With confidence weight set and call to install the digital benchmark model To determine the geometric constraint set Preload constraint set Fastener group stage sequence table and the set of discrete thicknesses of gaskets And based on the current installation stage The set of permissible control channels is selected from the four stages of "lifting and positioning control, alignment control, initial tightening control, and final tightening control" to form the decision variables for this round. and The component order is consistent with the control input increment vector in S4, including the push amount. , pullback amount Temporary support lifting capacity Incremental vector of gasket thickness combination With the tightening increment vector The minimum step size for both pushing and pulling back is 1mm, and the stroke boundary for a single movement is... The minimum step size for temporary support lifting is 0.5mm, and the single-action stroke boundary is... The temporary support bearing capacity is set at 150kN and is written into the system equipment parameter table as a hard constraint. The value of each component is limited to: The finite sum of the inner elements, which allows negative values ​​to represent subtraction pads. Each component corresponds to Only one fastener group is allowed to have a non-zero value during the initial tightening and final tightening control phases, and the value step is the design value of the preload of that fastener group. The maximum single-stage loading increment is set to the design value of the preload of the fastener group. ;

[0116] by For candidates To perform response prediction, specifically, first calculate the predicted installation response increment vector. Then The components corresponding to the six degrees of freedom deviation changes are superimposed on Obtain the predicted six-degree-of-freedom installation deviation vector and will The component corresponding to the change in preload is superimposed on Obtain the predicted fastener preload state vector ,in and Used to participate in both target evaluation and constraint determination;

[0117] A constrained control model is established, and the optimal control input is solved on a finite candidate set. The constrained control model is written as follows:

[0118]

[0119] This represents the optimal control input increment vector obtained in this round of solving and that will be implemented. This represents the independent variable that minimizes the objective function within the parentheses. Indicates the installation phase The feasible region is determined by the set of geometric constraints. With preload constraint set Together with the equipment parameter table, it is limited. The diagonal weighted matrix represents the six-degree-of-freedom installation bias, with its diagonal elements composed of the confidence weights of the observation factors associated with the six-degree-of-freedom bias. The aggregated components are then set to zero for those marked as the target elimination factor. The diagonal weighted matrix represents the fastener preload deviation, and its diagonal elements are composed of confidence weights associated with the preload observations. The fasteners identified by aggregation are then zeroed out, and the fasteners marked as the target rejection factor are also zeroed out. Indicate execution The predicted six-degree-of-freedom installation deviation vector is then generated. Represents the L2 norm, Indicate execution The predicted fastener preload state vector is then obtained. Indicates the set of preload constraints The vector of preload design values ​​obtained by assembling according to the fastener number order Table showing the preload dispersion index and in order of fastener group stages For each symmetrical fastening group, calculate the ratio of "the difference between the maximum and minimum preload force within the group divided by the design preload force value for that group", and take the maximum value of each ratio. Indicates the weight of the target term for preload dispersion and takes This indicates the weight of the target term for regulating the change in input and taking... ;

[0120] feasible region The geometric constraint part requires predicting the six-degree-of-freedom installation deviation vector. The axial displacement deviation, lateral displacement deviation, vertical displacement deviation, rotation angle deviation around the axis, and rotation angle deviation around the lateral and vertical angles fall into the following categories respectively: Within the corresponding tolerance boundary, the angle between the end face fitting gap and the end face normal must be satisfied. The end face plane constraint parameters are written and the angle between the hole misalignment and the hole axis satisfies the following conditions. Write the hole position coincidence constraint parameters;

[0121] feasible region The preload constraint section requires prediction of the fastener preload state vector. The preload value of each fastener falls into The preload force is allowed to be within the upper and lower bounds of the specified range, and the preload force dispersion index is required. No more than The preload dispersion limit is written;

[0122] feasible region Equipment and process constraints requirements To satisfy the above travel boundaries and minimum step size, the temporary support must not exceed the load-bearing boundary after the lifting / lowering amount is executed. Only by The inner thickness is superimposed and its corresponding position set is The spacers to be written must be placed in the same position, requiring Non-zero components only appear in The permitted tightening stages and the tightening sequence within the tightening group are the same as This ensures that the fastening sequence constraint is treated as a hard constraint.

[0123] Press the fasteners The system is divided into a symmetrical tightening group and a re-test tightening group, and graded loading parameters and re-test points are generated accordingly. The symmetrical tightening group is used to control the consistency within the group and the symmetry between groups, while the re-test tightening group is used to force a re-test and verify the preload dispersion after each tightening stage. For each symmetrical tightening group, the preload deviation within the group and the preload deviation between groups are calculated, and the difference between the remaining preload that does not reach the design value is used as the criterion. As the next tightening stage loading increment for this group, and simultaneously satisfying the loading increment step as follows: With the maximum "Identify the group" Select candidate values ​​and fix the holding time at 60s to allow the preload reading to stabilize before retesting;

[0124] When based on The predicted loading increment during this fastening phase will lead to Any component crosses When dealing with the tolerance boundary, the loading increment of the tightening stage is backed down by one step, and non-zero candidates for the alignment control input channel are added to the same round of candidate set, thereby improving the solution obtained. It may include both fastening loading and alignment actions, or only alignment actions;

[0125] The objective function value is calculated by traversing the finite candidate set and eliminating infeasible candidates. Candidates satisfying the convergence criterion are then selected as... The convergence criterion is determined by the "predicted six-degree-of-freedom installation deviation vector". Each component enters the corresponding tolerance boundary and predicts the fastener preload state vector. Each component enters the corresponding allowable preload range and "Not exceeding the dispersion limit" and "The objective function decreases by no less than the previous objective function" "Combined as a whole;

[0126] Will The code decodes the commands and outputs the pushing direction, pulling direction, step size, shim thickness combination, tightening sequence, graded loading parameters, and retest points. The pushing direction and pulling direction are determined by... and The sign is determined and the step size is taken as its absolute value; the gasket thickness combination is determined by... Output the gaskets one by one according to their placement, with each position outputting "Add Thickness or Decrease Thickness" and its corresponding thickness value. The tightening sequence is as follows: The set of fastening groups allowed in this stage and the order within the group are directly output. The graded loading parameters are composed of the loading increment, holding time of 60s and retest points for each fastening group. The retest points include the damper body pose target, the pose targets of the two end connectors, the reference target of the connecting beam connection node, and all fastener preload acquisition points in the fastening group in this stage, and additionally include the measurement points marked by S4 as the factors to be retested.

[0127] The execution and feedback of control commands are broken down into four action sequences according to the installation stage: hoisting and positioning control, alignment control, initial tightening control, and final tightening control. After each action in jacking, pulling back, temporary support lifting and lowering, shim addition and removal, and graded tightening, the retest data is immediately acquired according to the retest points and packaged into new unified installation observation data. ,Will As a new observation factor, it is fed into S3 and merged with the previous round of installation status factor diagram to update. and When each component of the updated six-degree-of-freedom installation deviation vector is within the corresponding tolerance boundary and the estimated value of the preload force of each fastener is within the corresponding allowable range of preload force and the preload force dispersion does not exceed the dispersion limit, an installation closed loop completion mark is generated and the loop is terminated. Otherwise, return to S3 and S4 to enter the next round of incremental optimization, self-correction and constraint control solution until the mark generation is completed.

[0128] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0129] This invention, through the continuous feedback relationship between the installation of a digital benchmark model, unified installation observation data, installation state factor diagram, and constraint control model, enables the measurement results of each stage of hoisting, positioning, initial tightening, and final tightening to be transformed into an iteratively optimized installation state. It also establishes a correspondence between actions such as jacking, pulling back, temporary support lifting, shim adjustment, and graded tightening and the six-degree-of-freedom deviation and preload state, thereby achieving a closed-loop control effect that matches the quality control problem of the coupling beam damper installation.

[0130] This invention further sets up a self-calibration mechanism for deviation-preload coupling sensitivity and a confidence weight generation mechanism. The control inputs already executed on site and the installation response obtained from the retest are continuously used to correct the coupling sensitivity matrix. Abnormal measurements or responses are downweighted, limited, or retested, so that the control model no longer depends on the preset parameters, but can be updated according to the actual constraints on site, thereby better ensuring the consistency of installation accuracy and preload.

Claims

1. A real-time monitoring and control method for the installation process of a coupling beam damper, characterized in that, include: S1. Based on the damper design axis, the mounting surfaces of the two end connectors, bolt hole positions, elevation reference, installation tolerance, preload design value, and fastening sequence constraints, establish an installation digital reference model, generating an installation reference coordinate system, geometric constraint set, and preload constraint set; S2. Collect the pose data and preload data of the damper body, two end connectors, connecting beam connection nodes, and each fastener, and transform them to the installation reference coordinate system to obtain unified installation observation data; S3. Construct an installation state factor diagram using the unified installation observation data, geometric constraint set, and preload constraint set, and incrementally optimize to obtain a six-degree-of-freedom installation deviation vector and a fastener preload state vector; S4. Based on the already executed... The coupling sensitivity matrix is ​​recursively corrected based on the changes in deviation and preload obtained from the line control amount and retest, and confidence weights are generated based on the sensor residuals; S5, the constraint control model is solved based on the installation stage, the six-degree-of-freedom installation deviation vector, the fastener preload state vector, confidence weights, the coupling sensitivity matrix, the geometric constraint set, and the preload constraint set. Control commands including the pushing direction, pulling direction, step amount, gasket thickness combination, fastening sequence, graded loading parameters, and retest points are generated and executed. The retest data is fed back to the installation state factor diagram until the six-degree-of-freedom installation deviation vector and the fastener preload state vector satisfy the geometric constraint set and the preload constraint set.

2. The real-time monitoring and control method for the installation process of the coupling beam damper according to claim 1, characterized in that, S1 includes: establishing an installation reference coordinate system with the construction measurement benchmark of the coupling beam as the coordinate reference; The damper design axis and the mounting surfaces of the connectors at both ends are converted into axial constraints and end face plane constraints, respectively; the bolt hole positions, elevation datum, and installation tolerances are converted into hole position coincidence constraints, elevation constraints, and six-degree-of-freedom tolerance boundaries; the preload design value, preload allowable deviation, and preload dispersion limit are converted into a preload constraint set; and the fastening sequence constraints are converted into a stage sequence table of fastener groups to obtain the installation digital datum model.

3. The real-time monitoring and control method for the installation process of the coupling beam damper according to claim 1, characterized in that, S2 includes: acquiring raw measurement data through pose targets set on the damper body and the two end connectors, a reference target set on the connecting beam node, and a preload acquisition unit for fasteners; aligning the raw measurement data with timestamps and calibrating the equipment coordinates; converting the pose data into the poses of the damper body, the two end connectors, and the connecting beam node using the external parameter transformation between the installation reference coordinate system and the coordinate systems of each acquisition device; and mapping the preload acquisition data to preload observation values ​​according to the fastener number to form unified installation observation data.

4. The real-time monitoring and control method for the installation process of the coupling beam damper according to claim 1, characterized in that, S3 includes: using the damper body pose, the poses of the two end connectors, the poses of the connecting beam connection nodes, and the preload of each fastener as variable nodes; using pose observation, end face fit, coaxial axis, hole position correspondence, elevation, fastening sequence, and preload design constraints as factor nodes; setting residual weights for the factor nodes according to confidence weights or preset measurement weights; using the previous round of optimization results as a priori, incrementally solving the installation state factor diagram, and outputting a six-degree-of-freedom installation deviation vector composed of axial displacement deviation, lateral displacement deviation, vertical displacement deviation, rotation angle deviation around the axis, rotation angle deviation around the lateral angle deviation, and rotation angle deviation around the vertical angle deviation, as well as a fastener preload state vector composed of the estimated values ​​of each fastener preload.

5. The real-time monitoring and control method for the installation process of the coupling beam damper according to claim 1, characterized in that, S4 includes: encoding the executed jacking amount, pullback amount, temporary support lifting amount, shim thickness combination, and fastening increment into a control input increment vector; encoding the difference between the six-degree-of-freedom installation deviation vector and the difference between the fastener preload state vector before and after the retest into an installation response increment vector; using the control input increment vector and the installation response increment vector to form an update sample, updating the coupling sensitivity matrix using recursive least squares with a forgetting factor, and constraining the update result through element amplitude limiting and sign consistency verification; inputting the updated coupling sensitivity matrix into the constraint control model of step S5 to predict the impact of the next round of control commands on the six-degree-of-freedom installation deviation vector and the fastener preload state vector.

6. The real-time monitoring and control method for the installation process of the coupling beam damper according to claim 1, characterized in that, S5 includes: establishing a constraint control model with six-degree-of-freedom installation deviation weighted amount, preload deviation weighted amount, preload dispersion, and control input change as target terms; using geometric constraint set, preload constraint set, jacking mechanism stroke boundary, temporary support bearing boundary, shim discrete thickness set, and fastening sequence table as constraint conditions; predicting the six-degree-of-freedom installation deviation vector and fastener preload state vector corresponding to candidate control commands based on the coupling sensitivity matrix, and adjusting the target term weights according to confidence weights; selecting control commands whose target term values ​​satisfy the convergence criterion from candidate control commands that meet the constraint conditions, and outputting the jacking direction, pullback direction, step amount, shim thickness combination, fastening sequence, graded loading parameters, and retest points.

7. The real-time monitoring and control method for the installation process of the coupling beam damper according to claim 6, characterized in that, The execution and feedback of control commands include: decomposing control commands into hoisting and positioning control, alignment control, initial tightening control, and final tightening control according to the installation stage; after each action in jacking, pulling back, temporary support lifting, shim addition / reduction, and graded tightening, acquiring retest data according to the retest points; using the retest data as a new observation factor and merging it with the previous round of installation state factor diagram; when each component of the six-degree-of-freedom installation deviation vector is within the corresponding tolerance boundary, the estimated value of each fastener preload is within the corresponding preload range, and the preload dispersion does not exceed the dispersion limit, generating an installation closed-loop completion mark; if not satisfied, returning the new observation factor to S3 and continuing the next round of optimization and control.

8. The real-time monitoring and control method for the installation process of the coupling beam damper according to claim 5, characterized in that, The confidence weights are generated by normalizing the pose observation residual, the preload observation residual, the change in the same measurement point at adjacent times, and the observation difference of different sensors at the same measurement point to obtain the measurement residual component, the time consistency component, and the sensor consistency component. The basic confidence weights of each observation factor are generated based on the preset noise threshold and the residual component mapping table. When the basic confidence weight of any observation factor is lower than the retest trigger threshold, the observation factor is marked as a factor to be retested and its residual weight when entering S3 is restricted; when it is still lower than the retest trigger threshold after retesting, the observation factor is removed from the target item of the current constraint control model, and the confidence weight is used to adjust the target item weight.

9. The real-time monitoring and control method for the installation process of the coupling beam damper according to claim 6, characterized in that, The fastening sequence and graded loading parameters are determined as follows: Fasteners are divided into symmetrical fastening groups and retest fastening groups based on the fastening sequence table in the installation digital reference model; the preload deviation within and between groups of each symmetrical fastening group is calculated based on the fastener preload state vector; the loading increment, holding time, and retest point for each fastening stage are determined according to the preload design value, intra-group preload deviation, inter-group preload deviation, and coupling sensitivity matrix; when the predicted loading increment causes any six-degree-of-freedom installation deviation component to exceed the tolerance boundary, the loading increment of the corresponding fastening stage is reduced, and a correction control input is added to the next round of constraint control model.

10. A real-time monitoring and control system for the installation process of a coupling beam damper, used to execute the real-time monitoring and control method for the installation process of a coupling beam damper as described in any one of claims 1 to 9, characterized in that, include: The benchmark model establishment module is used to establish an installation digital benchmark model based on the damper design axis, the mounting surfaces of the connectors at both ends, the bolt hole positions, the elevation benchmark, the installation tolerance, the design value of the preload, and the fastening sequence constraints, and to generate the installation benchmark coordinate system, the geometric constraint set, and the preload constraint set; the installation data acquisition module is used to collect the position and orientation data and preload data of the damper body, the connectors at both ends, the connecting beam connection nodes, and each fastener, and to output unified installation observation data; The factor graph optimization module is used to construct the installation state factor graph and incrementally optimize it to obtain the six-degree-of-freedom installation deviation vector and the fastener preload state vector; the self-calibration and confidence evaluation module is used to recursively correct the coupling sensitivity matrix and generate confidence weights based on the sensor residuals. The constraint control closed-loop module is used to generate and execute control commands based on the installation stage, the six-degree-of-freedom installation deviation vector, the fastener preload state vector, the confidence weight, the coupling sensitivity matrix, the geometric constraint set, and the preload constraint set, and feeds back the retest data to the installation state factor diagram.

Images