A method and related apparatus for analyzing the omnidirectional deformation state of GIL expansion joints

Abstract

This invention belongs to the field of gas-insulated metal-enclosed transmission lines (GILs), and discloses a method and related device for omnidirectional deformation state analysis of GIL expansion joints. The method involves real-time acquisition of relative distance data between measuring points on both sides of the expansion joint flanges; establishing a three-dimensional relative coordinate system with the center of either flange as the origin; introducing 6-DOF pose parameters based on the geometric layout of the measuring points; constructing a deformation motion model of the expansion joint based on rigid body kinematics principles; and constructing an objective function for the model with the 6-DOF pose parameters as variables. A nonlinear least squares optimization algorithm is used to solve the model to obtain the 6-DOF deformation state; dynamic smoothing optimization is performed to form a real-time spatial deformation state sequence of the expansion joint; the pre-trained model is input to output the spatial deformation state prediction results; and a structural health assessment report and early warning information are generated. This method achieves omnidirectional deformation monitoring, improves the accuracy of state assessment by integrating multi-dimensional parameters through deep data analysis, reduces early warning lag and false alarms / missed alarms, and improves system reliability and operation and maintenance efficiency.

Description

Technical Field

[0001] This invention belongs to the technical field of gas-insulated metal-enclosed transmission lines, specifically to the field of monitoring expansion joints of gas-insulated metal-enclosed transmission lines, and particularly relates to a method and related device for analyzing the omnidirectional deformation state of GIL expansion joints. Background Technology

[0002] Gas-insulated metal-enclosed transmission lines (GILs) play an indispensable role in clean energy projects. Some special regions have extremely complex geological conditions, characterized by complex geological structures, high seismic risk, the presence of deep riverbed overburden, and high ground temperatures. Under these complex conditions, GIL lines must traverse areas with variable geological conditions and experience significant diurnal and seasonal temperature differences. They face complex operating conditions including long distances, significant elevation differences, and geological subsidence. Expansion joints, as key components, absorb stress through the elastic deformation of their bellows structure to protect the rigid shell of the GIL. However, expansion joints are prone to exceeding deformation limits due to concentrated vertical forces, temperature changes, and surface subsidence, leading to issues such as misalignment and torsion. This can result in serious faults such as seal failure, partial discharge, and even insulation breakdown. Therefore, condition monitoring of GIL expansion joints is an urgent practical need.

[0003] Current monitoring methods for GIL expansion joints have many shortcomings. Traditional manual measurement and inspection methods are labor-intensive, lack real-time performance, and are costly. The mechanical reading scales commonly used in existing substation GIL expansion joints do not have recording functions, requiring manual on-site recording, which is not only time-consuming and inefficient but also prone to errors due to untimely recording or misreading. Furthermore, it is difficult to capture transient anomalies and long-term trend deterioration. Existing digital monitoring technologies also have significant drawbacks: first, their technological maturity is insufficient, resulting in difficulties in installation, inadequate measurement accuracy, and poor long-term operational reliability; second, some technologies rely on expensive sensors, keeping costs high; and third, there is insufficient in-depth analysis of measurement data and assessment of expansion joint condition, failing to realize the deeper value of the data. Specifically, most existing technologies lack omnidirectional deformation monitoring capabilities, monitoring only single-dimensional deformations such as axial deformation. They cannot comprehensively measure and identify multi-dimensional deformations of GIL expansion joints, such as axial expansion, radial offset, and angular misalignment. Data analysis depth and alarm mechanisms are limited; data processing remains at the level of simple acquisition and uploading, lacking multi-source data fusion and deep feature mining. Alarm logic relies solely on axial deformation exceeding a threshold, failing to comprehensively assess equipment status using multi-dimensional parameters. Trend prediction and early warning are lagging, unable to predict future deformation trends, hindering proactive maintenance. Furthermore, most existing technologies merely convert manual judgment standards into automated programs, unable to analyze complex deformation situations like experts, easily leading to delayed warnings, false alarms, or missed alarms.

[0004] In summary, existing analytical methods cannot achieve comprehensive monitoring, in-depth data analysis, and accurate trend prediction of multidimensional deformation of GIL expansion joints under complex geological conditions. They are insufficient to meet the proactive operation and maintenance needs of GIL expansion joints and cannot effectively avoid serious failures caused by excessive deformation of expansion joints. Summary of the Invention

[0005] This invention provides a method and related device for omnidirectional deformation state analysis of GIL expansion joints. This method can effectively solve the problem that existing analysis methods cannot achieve comprehensive monitoring, in-depth data analysis and accurate trend prediction of multidimensional deformation of GIL expansion joints under complex geological conditions. It can meet the active operation and maintenance needs of GIL expansion joints and effectively avoid serious failures caused by excessive deformation of expansion joints.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A method for analyzing the omnidirectional deformation state of GIL expansion joints includes: Real-time acquisition of relative distance data between corresponding measuring points on both sides of the target expansion joint flange; A three-dimensional relative coordinate system is established with one of the flange centers on the left and right sides of the target expansion joint as the origin. The initial coordinates are set in combination with the known geometric layout of multiple measuring points on the flange. Six degrees of freedom pose parameters are introduced, and the expansion joint deformation motion model is constructed based on the rigid body kinematics principle. An objective function with six degrees of freedom pose parameters as variables is constructed for the expansion joint deformation motion model. Based on the relative distance data between multiple measurement points at the current time and historical time, a nonlinear least squares optimization algorithm is used to solve the objective function of the expansion joint deformation motion model to obtain the 6-DOF deformation state of the target expansion joint; wherein, the 6-DOF deformation state is represented by 6-DOF pose parameters. The 6-DOF deformation state of the target expansion joint is dynamically smoothed and optimized to obtain the real-time spatial deformation state sequence of the expansion joint. The real-time spatial deformation state sequence of the expansion joint is input into a pre-trained spatial deformation state prediction model to output the spatial deformation state prediction result of the target expansion joint; wherein, the spatial deformation state prediction model is trained based on historical deformation time series data; Based on the predicted results of the 6-DOF deformation state and spatial deformation state of the target expansion joint, a structural health assessment report and early warning information are generated.

[0007] Furthermore, the real-time acquisition of relative distance data between corresponding measuring points on both sides of the target expansion joint flange includes: Multiple displacement sensors are pre-arranged symmetrically on the circumference of the flanges on both sides of the target expansion joint; The relative distance data between corresponding measuring points on both sides of the flange of the target expansion joint is collected in real time by displacement sensors to obtain the original distance sequence. The original distance sequence is preprocessed by filtering, and the filtered distance sequence is output for solving the objective function of the extensor deformation motion model.

[0008] Furthermore, a three-dimensional relative coordinate system is established with one of the flange centers on the left and right sides of the target expansion joint as the origin. Initial coordinates are set based on the known geometric layout of multiple measuring points on the flange. Six degrees of freedom pose parameters are introduced, and a deformation motion model of the expansion joint is constructed based on the principles of rigid body kinematics, including: A three-dimensional relative coordinate system is established with the center of the left flange of the target expansion joint as the origin; the construction principle of the coordinate system with the center of the right flange of the target expansion joint as the origin is the same as that of the left flange. The relative deformation state of the expansion joint is described as a 6-DOF variable: translation vector. Indicates to respectively Translation distance in direction, rotation vector Indicates to respectively and The angle of the axis, based on a given rotation vector. Constructing a rotation matrix using Rodriguez's formula :

[0009] In the formula, Represents the identity matrix. Indicates the rotation angle, used to show the extent of rotation of the right flange relative to the left flange. ; Represents the rotation vector The antisymmetric matrix; After deformation occurs, according to the rotation matrix Calculate the right flange number The theoretical location of each measuring point is Then, its current pose corresponds to the point on the left. Theoretical distance for:

[0010] In the formula, This indicates the three-dimensional coordinates of the current measuring point on the right flange; Indicates the current left flange and The corresponding three-dimensional coordinates of the measuring point; Represents Euclidean distance; The objective function for constructing the deformation motion model of the extension joint based on the principles of rigid body kinematics is as follows:

[0011] In the formula, This indicates the actual distance measured by the displacement sensor at each measuring point; For torsional motion, the regularization coefficient is denoted as . is the regularization coefficient for shear motion; This represents the pose parameters.

[0012] Furthermore, based on the relative distance data between multiple measurement points at the current and historical times, a nonlinear least squares optimization algorithm is used to solve the objective function of the expansion joint deformation motion model to obtain the 6-DOF deformation state of the target expansion joint, including: A nonlinear least squares optimization method is employed, using a vector of 6-DOF deformation parameters. To optimize the variables, the parameters are iteratively adjusted to ensure the theoretical calculated distance at each measuring point is optimized. Compared with the measured distance The objective function minimizes the residuals between the two values; in each iteration, the parameters are updated based on the current residuals and the Jacobian matrix until the objective function converges to a preset threshold, resulting in a 6-DOF vector after iterative convergence. This allows for the quantitative determination of the deformation state. The converged rotation vector Transformed into a rotation matrix using Rodriguez's formula Combined with translation vector The six-degree-of-freedom deformation state of the target expansion joint is constituted; based on the six-degree-of-freedom deformation state, the current pose of the flanges on both sides of the target expansion joint is determined.

[0013] Furthermore, the dynamic smoothing and optimization of the 6-DOF deformation state of the target expansion joint to obtain the real-time spatial deformation state sequence of the expansion joint includes: Predict the prior state at the current moment based on the posterior state at a historical moment; Using the 6-DOF deformation state of the target expansion joint as the observation input, the Kalman gain is dynamically calculated, and the prediction and observation information are fused to generate a corrected posterior state estimate. The corrected posterior state estimate is output as the real-time spatial deformation state sequence of the expansion joint to achieve dynamic smoothing and optimization of the 6-DOF deformation state of the target expansion joint.

[0014] Furthermore, before inputting the real-time spatial deformation state sequence of the expansion joint into the pre-trained spatial deformation state prediction model to output the spatial deformation state prediction result of the target expansion joint, the method further includes: constructing and training the spatial deformation state prediction model, the specific steps of which are as follows: Supervised learning samples are constructed using a sliding time window approach, wherein the supervised learning samples include historical deformation time series data and the corresponding deformation states at several future moments; The pre-constructed initial model for predicting spatial deformation state is trained using supervised learning samples, and the trained spatial deformation state prediction model is output; wherein, the base model of the initial model for predicting spatial deformation state adopts a long short-term memory network model.

[0015] Furthermore, based on the predicted results of the 6-DOF deformation state and spatial deformation state of the target expansion joint, a structural health assessment report and early warning information are generated, including: A two-dimensional, multi-level early warning standard covering both cumulative deformation and deformation rate is established, divided into three levels: Normal monitoring level: Deformation is within the allowable range, and the system records normally; Trend Deviation Level: Deformation or rate exceeds the normal threshold but does not reach a dangerous level, triggering a low-risk alarm; Critical risk level: The predicted deformation or rate is close to the structural failure limit, triggering a high-risk alarm; The predicted results of the 6-DOF deformation state and spatial deformation state of the target expansion joint are compared in real time with the deformation threshold and deformation rate threshold to generate early warning information of the corresponding level. A structural health assessment report is generated based on the predicted results of the 6-DOF deformation state and spatial deformation state of the target expansion joint.

[0016] A GIL (Gas Inertia Joint) omnidirectional deformation state analysis system, comprising: The data acquisition module is used to collect the relative distance data between corresponding measuring points on both sides of the target expansion joint in real time. The model building module is used to establish a three-dimensional relative coordinate system with one of the flange centers on the left and right sides of the target expansion joint as the origin. It sets the initial coordinates based on the known geometric layout of multiple measuring points on the flange, introduces 6-DOF pose parameters, and constructs the expansion joint deformation motion model based on the rigid body kinematics principle. It also constructs an objective function for the expansion joint deformation motion model with 6-DOF pose parameters as variables. The morphology solution module is used to solve the objective function of the expansion joint deformation motion model based on the relative distance data between multiple measurement points at the current time and historical time, using a nonlinear least squares optimization algorithm, so as to obtain the 6-DOF deformation state of the target expansion joint; wherein, the 6-DOF deformation state is represented by 6-DOF pose parameters; The data correction module is used to dynamically smooth and optimize the 6-DOF deformation state of the target expansion joint in order to obtain the real-time spatial deformation state sequence of the expansion joint. The morphology prediction module is used to input the real-time spatial deformation state sequence of the expansion joint into a pre-trained spatial deformation state prediction model to output the spatial deformation state prediction result of the target expansion joint; wherein, the spatial deformation state prediction model is trained based on historical deformation time series data; The early warning module is used to generate structural health assessment reports and early warning information based on the prediction results of the 6-DOF deformation state and spatial deformation state of the target expansion joint.

[0017] A device for analyzing the omnidirectional deformation state of GIL expansion joints, comprising: Memory, used to store computer programs; A processor is used to implement the above-described GIL expansion joint omnidirectional deformation state analysis method when executing the computer program.

[0018] A computer-readable storage medium storing a computer program, which, when executed by a processor, is used to implement the above-described method for analyzing the omnidirectional deformation state of a GIL (Glass Inlet Joint).

[0019] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a method for analyzing the omnidirectional deformation state of GIL (Gas Inertial Joint) expansion joints. First, the relative distance data between measuring points on both sides of the expansion joint is collected in real time. A three-dimensional relative coordinate system is established with the center of either flange as the origin. Six-DOF (degree-of-freedom) pose parameters are introduced based on the geometric layout of the measuring points. A deformation motion model of the expansion joint is constructed based on the principles of rigid body kinematics, and an objective function is constructed for it, with the six-DOF pose parameters as variables. A nonlinear least squares optimization algorithm is used to solve the model to obtain the six-DOF deformation state. Dynamic smoothing optimization is performed to form a real-time spatial deformation state sequence of the expansion joint. The pre-trained model is input to output the spatial deformation state prediction results. Finally, a structural health assessment report and early warning information are generated. This method treats the expansion joint as a rigid body through a rigid body kinematic model, comprehensively capturing axial, radial, and angular deformation using multi-point data and six-DOF parameters. The optimization algorithm handles measurement errors, dynamically smooths to eliminate noise, and ensures data stability. The pre-trained model mines features based on historical time-series data to achieve trend prediction. This method enables omnidirectional deformation monitoring, overcoming the problems of poor real-time performance and high cost of traditional methods. By integrating multi-dimensional parameters through deep data analysis, it improves the accuracy of status assessment and avoids the limitations of single threshold alarms. It supports trend prediction to promote proactive operation and maintenance, reduces early warning lag and false alarms and missed alarms, and improves system reliability and operation and maintenance efficiency. Attached Figure Description

[0020] Figure 1 A flowchart illustrating the implementation of an omnidirectional deformation state analysis method for GIL expansion joints provided in this embodiment of the invention; Figure 2 A structural block diagram of a GIL expansion joint omnidirectional deformation state analysis system provided in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the installation position of the target expansion joint sensor and the method of establishing coordinate axes provided in an embodiment of the present invention; Figure 4This is a core flowchart of a method for analyzing the omnidirectional deformation state of a GIL expansion joint, provided in an embodiment of the present invention. Figure 5 This is a schematic diagram of a GIL expansion joint omnidirectional deformation state analysis system provided in an embodiment of the present invention. Detailed Implementation

[0021] To further understand the content of this invention, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments are merely illustrative and not limiting of the invention.

[0022] The technical terms used in this invention are explained below: GIL: an abbreviation for Gas-Insulated Metal-Enclosed Transmission Line, also commonly known as gas-insulated pipeline transmission line. It is a high-voltage transmission technology consisting of a central conductor, a coaxial grounded metal casing, and an insulating gas (usually sulfur hexafluoride SF6 or a mixture thereof) between them. The entire structure is sealed within a metal pipe filled with a high-insulation-strength gas, offering advantages such as large transmission capacity, low loss, low electromagnetic interference, high safety, and small footprint. GIL is commonly used in applications with limited space, sensitive environments, or high reliability requirements, such as urban underground power transmission, hydroelectric power station outgoing lines, internal connections in nuclear power plants, or special routes crossing rivers, tunnels, etc. Due to its fully enclosed structure, GIL is virtually unaffected by external weather, pollution, or mechanical damage, but it is more expensive and its installation and maintenance are relatively complex.

[0023] A displacement sensor is a device used to measure changes in the position (i.e., displacement) of an object. It converts mechanical displacement into an electrical signal output, thereby enabling the monitoring, recording, or control of displacement. Displacement can be linear (linear motion) or angular (rotational motion), therefore displacement sensors are divided into two main categories: linear displacement sensors and angular displacement sensors.

[0024] Mathematical modeling refers to using mathematical language, symbols, formulas, and logical structures to describe problems or phenomena in the real world, thereby analyzing, predicting, optimizing, or controlling the behavior of the system. It serves as a bridge connecting practical problems with mathematical theory and is widely used in science, engineering, and many other fields.

[0025] As mentioned in the background section, current monitoring and analysis methods suffer from insufficient omnidirectional deformation monitoring capabilities, limited data analysis depth and alarm mechanisms, and lagging trend prediction and early warning.

[0026] To address the aforementioned issues, this embodiment provides a method for analyzing the omnidirectional deformation state of GIL expansion joints. This method enables real-time, automatic, and high-precision monitoring, analysis, and early warning of the omnidirectional spatial deformation state of expansion joints, thereby improving the reliability and intelligent operation and maintenance level of long-distance GIL transmission equipment.

[0027] like Figure 4 As shown, this embodiment provides a method for analyzing the omnidirectional deformation state of a GIL (Gas Inertial Joint) expansion joint, including: Real-time acquisition of relative distance data between corresponding measuring points on both sides of the target expansion joint flange; A three-dimensional relative coordinate system is established with one of the flange centers on the left and right sides of the target expansion joint as the origin. The initial coordinates are set in combination with the known geometric layout of multiple measuring points on the flange. Six degrees of freedom pose parameters are introduced, and the expansion joint deformation motion model is constructed based on the rigid body kinematics principle. An objective function with six degrees of freedom pose parameters as variables is constructed for the expansion joint deformation motion model. Based on the relative distance data between multiple measurement points at the current time and historical time, a nonlinear least squares optimization algorithm is used to solve the objective function of the expansion joint deformation motion model to obtain the 6-DOF deformation state of the target expansion joint; wherein, the 6-DOF deformation state is represented by 6-DOF pose parameters. The 6-DOF deformation state of the target expansion joint is dynamically smoothed and optimized to obtain the real-time spatial deformation state sequence of the expansion joint. The real-time spatial deformation state sequence of the expansion joint is input into a pre-trained spatial deformation state prediction model to output the spatial deformation state prediction result of the target expansion joint; wherein, the spatial deformation state prediction model is trained based on historical deformation time series data; Based on the predicted results of the 6-DOF deformation state and spatial deformation state of the target expansion joint, a structural health assessment report and early warning information are generated.

[0028] The prediction method provided in this embodiment will be further explained below with reference to the accompanying drawings: like Figure 1As shown, this analysis method involves arranging displacement sensors at multiple measuring points on the circumference of the left and right flanges of the GIL expansion joint to collect distance data between corresponding points on both flanges in real time. The collected raw data is pre-processed by filtering to obtain a noise-reduced distance sequence. Then, taking the center of the left flange of the expansion joint as the origin, a three-dimensional relative coordinate system is established, and an objective function is constructed based on a 6-DOF spatial deformation mathematical model, transforming the spatial deformation problem of the expansion joint into a parameter optimization problem of the objective function; the principle of constructing the coordinate system with the center of the right flange of the expansion joint as the origin is the same as that on the left. Using the distance data from multiple measuring points at the current and historical times, the 6-DOF deformation state of the expansion joint is iteratively calculated. Subsequently, an extended Kalman filter (EKF) is used to dynamically filter and smooth this state to obtain a high-precision deformation estimation result. Furthermore, a deformation trend prediction model is constructed by integrating a Long Short-Term Memory (LSTM) network and pattern recognition technology. Based on the current deformation state and future evolution trend, the system generates a deformation analysis report for the expansion joint and triggers a tiered early warning mechanism according to preset thresholds to promptly prompt maintenance personnel to carry out maintenance work. The specific steps are as follows: Step S1: Data Acquisition and Preprocessing. Multiple displacement sensors are evenly arranged on the circumference of the left and right flanges of the target expansion joint to collect the relative distance data between corresponding measuring points on both flanges in real time; the raw distance data is filtered and preprocessed to suppress environmental noise and measurement interference, resulting in a noise-reduced multi-point distance data sequence.

[0029] Step S2: Mathematical Model Establishment. A three-dimensional relative coordinate system is established with the center of the left flange as the origin. Based on the known geometric layout of the sensors on the flange, the initial poses of the left and right flanges are defined. Based on the principle of rigid body kinematics, 6-DOF pose parameters (3 translational components and 3 rotational components) are introduced to construct a mathematical model of the expansion joint deformation (expansion joint deformation motion model). An objective function with pose parameters as variables is constructed, transforming the problem of estimating the spatial deformation state of the expansion joint into an optimization problem of solving this objective function.

[0030] Step S3: Spatial Deformation State Calculation. Based on the distance data of multiple measurement points at the current and historical times, the objective function is solved using the model established in Step S2 through a nonlinear least squares optimization algorithm to obtain the current 6-DOF deformation state of the expansion joint.

[0031] Step S4: Spatial Deformation State Correction. The preliminary deformation state output in Step S3 is dynamically smoothed and optimized. An Extended Kalman Filter (EKF) is used to fuse historical states with current measurements. The deformation state estimate is recursively updated through a prediction-correction mechanism to filter out abnormal fluctuations and output a continuous, smooth, and high-precision real-time spatial deformation state sequence of the expansion joint.

[0032] Step S5: Spatial Deformation State Prediction. A Long Short-Term Memory (LSTM) network and a deformation pattern recognition algorithm are integrated, and a dynamic prediction model (spatial deformation state prediction model) is trained using historical deformation time-series data to predict the future spatial deformation trend of multi-step expansion joints.

[0033] Step S6: Issue an early warning and generate a report. Based on the current 6-DOF deformation state output in Step S4 and the future evolution trend predicted in Step S5, the system automatically generates a structural health assessment report, covering historical changes, current status, and future risk predictions. It also triggers a corresponding tiered early warning mechanism based on preset safety thresholds, promptly prompting maintenance personnel to take repair measures.

[0034] For example, in step S1: First, define the sensor layout. For example... Figure 3 As shown, multiple displacement sensors (four in total) are symmetrically arranged on the circumference of the flanges on both sides of each GIL expansion joint, located directly above, in front, below, and behind, respectively, ensuring that the sensors on the left and right flanges correspond one-to-one in circumferential orientation. With this layout, if the GIL includes... If there are 1 expansion joint, then a total of 1000 expansion joints are needed. One displacement sensor.

[0035] Then, the relative distance data between the corresponding measuring points on the left and right flanges of the target expansion joint are collected in real time by the aforementioned sensors to form the original distance sequence; and the original data is filtered and preprocessed (such as low-pass filtering or wavelet denoising) to suppress environmental noise and measurement interference, so as to obtain denoised distance data with high signal-to-noise ratio for subsequent modeling and analysis.

[0036] For example, in step S2: This step, based on preprocessed distance data, constructs a joint deformation motion model in 3D space using 6-DOF pose parameters and establishes an objective function for state estimation. Specifically, it includes the following sub-steps: Step S21: Define the coordinate system and sensor layout. Establish a three-dimensional relative coordinate system with the center of the left flange of the target expansion joint as the origin: The shaft points to the right along the axial direction of the expansion joint. The axis is horizontal and forward. The axis is vertically upward. This coordinate system moves with the left flange, and its origin is not a fixed point in space. Let the initial distance between the left and right flanges be... (Free length of the bellows), the sensor is installed at a radius of On the circumference. For example... Figure 3 As shown, the four sensors are evenly distributed along the circumference (at intervals of...). The coordinates on the left flange are as follows: The initial coordinates of the corresponding right flange are .

[0037] Step S22: Construction of a 6-DOF deformation motion model. The relative deformation state of the expansion joint is described as a 6-DOF variable: translation vector. Indicates to respectively Translation distance in direction, rotation vector Indicates to respectively and The angle of the axis, given the rotation vector A rotation matrix is ​​established based on Rodrigues' formula. :

[0038] in, Represents the identity matrix. This indicates the rotation angle, specifically the degree of rotation of the right flange relative to the left flange. . Represents the rotation vector An antisymmetric matrix.

[0039] After deformation, according to the rotation matrix Calculate the right flange number The theoretical location of each measuring point is Then, its current pose corresponds to the point on the left. The theoretical distance is (using express):

[0040] in This represents the coordinates of the measuring point on the right flange. The current left flange and The coordinates of the corresponding measuring points are represented as follows: It represents Euclidean distance.

[0041] Step S23: Constructing the objective function. The objective function is the sum of squared residuals between the distances measured by the displacement sensors at each point and the distances calculated based on the rotation matrix in step S22. A regularization term is introduced to suppress overfitting.

[0042] in, This indicates the actual distance measured by the displacement sensor at each measuring point, such as... This indicates the measured distance of the first displacement sensor. This is the regularization coefficient for torsional motion, used to control the weight of torsional motion parameters in the optimization process, avoiding overfitting of the model to torsional deformation due to sensor noise or local abnormal data. This is the regularization coefficient for shear motion, which applies similar constraints to the shear motion parameters to prevent overfitting in shear deformation calculations.

[0043] For example, in step S3: Based on the distance data of the corresponding measuring points on the left and right flanges of the expansion joint collected at the current moment, the objective function constructed in step S2 is... The solution is obtained by iterative optimization, using the deformation state of the previous moment as the initial estimate, to obtain the high-precision 6-DOF deformation parameters at the current moment.

[0044] Step S31: Iterative optimization solution. A nonlinear least squares optimization method is used, with the vector consisting of 6 degrees of freedom deformation parameters. To optimize the variables, the parameters are iteratively adjusted to ensure the theoretical calculated distance at each measuring point is optimized. Compared with the measured distance Minimize the residuals between them. In each iteration, update the parameters based on the current residuals and the Jacobian matrix until the objective function converges to a preset threshold, yielding a 6-DOF vector after iterative convergence. This allows for the quantitative determination of the deformation state.

[0045] Step S32: Deformation state output. Output the converged rotation vector. Convert to a rotation matrix using Rodrigues' formula. Combined with translation vector This constitutes the current 6-DOF relative pose of the expansion joint. Based on this pose and the coordinate system of the left flange, the spatial coordinates of each measurement point on the right flange at the current moment can be calculated, thus completing the quantitative characterization of the deformation state.

[0046] For example, in step S4: First, a state-space model of the joint deformation is constructed, treating the 6-DOF deformation state as a system state variable that evolves over time. A state transition model describing its dynamic evolution and an observation model associated with the observed data are then established. Based on the statistical characteristics of historical operating data, filter parameters are rationally initialized, particularly by setting a larger observation noise covariance to reduce the impact of sudden interference or abnormal data on the estimation results and improve system robustness.

[0047] Then, a recursive prediction-correction mechanism is used for real-time estimation: first, the prior state at the current moment is predicted based on the posterior state at historical moments; then, the current 6-DOF deformation state output in step S3 is used as the observation input to dynamically calculate the Kalman gain, fuse the prediction and observation information, and generate a corrected posterior state estimate. Finally, a smooth, continuous, and noise-resistant optimal deformation state sequence is output for subsequent trend prediction and early warning analysis.

[0048] For example, in step S5: Based on the high-precision 6-DOF deformation state time series data output in step S4, a deep learning prediction model is constructed to realize online prediction of future multi-step deformation trends.

[0049] Step S51: Construction of temporal samples. Supervised learning samples are constructed using a sliding time window approach. Each sample contains a historical deformation sequence (input) and the deformation state at several future time points (target output). The training set and validation set are divided to ensure temporal continuity and avoid data leakage.

[0050] Step S52: Model Training and Online Prediction. A Long Short-Term Memory (LSTM) network model is constructed, directly inputting a 6-DOF pose sequence to automatically learn deformation evolution patterns. Offline training is used to converge the validation error to a preset threshold, forming a prediction engine. During online operation, the model receives the real-time output from step S4 and continuously predicts future deformation trends (such as deformation rate and cumulative displacement increment). The system supports periodic retraining of the model using newly accumulated data to adapt to the slow drift of deformation characteristics over long-term operation.

[0051] For example, in step S6: Based on the design specifications and safe operation requirements of GIL equipment, a two-dimensional, multi-level early warning standard covering "cumulative deformation" and "deformation rate" is established, divided into three levels: Routine monitoring: Deformation is within the allowable range, and the system records normally; Trend deviation: Deformation or rate exceeds the normal threshold but does not reach a dangerous level, triggering a low-risk alarm; Critical risk: When the predicted deformation or rate approaches the structural failure limit, a high-risk alarm is triggered.

[0052] The system compares the predicted maximum cumulative deformation and maximum deformation rate from step S5 with the aforementioned thresholds in real time, and introduces a continuity verification mechanism (such as continuous...). (An alarm is only triggered when the predicted point exceeds the limit), effectively suppressing false alarms caused by occasional fluctuations.

[0053] Differentiated responses are implemented based on risk levels: when a trend deviates, the event is recorded and a notification is sent; when a critical risk occurs, an audible and visual alarm is immediately triggered, an emergency notification is sent to maintenance personnel, and the location of the specific expansion joint is pinpointed. Simultaneously, the system automatically generates a structural health assessment report, integrating historical data, current status, and future risk predictions, and archives warning events to build a digital management archive for the entire equipment lifecycle.

[0054] To implement the above analysis method, such as Figure 2As shown in the figure, this embodiment also provides a GIL expansion joint omnidirectional deformation state analysis system, including: a data acquisition module, a data preprocessing module, a deformation state analysis module, a deformation state correction module, a deformation trend analysis module, and an early warning module connected in sequence to form a complete data processing flow, realizing full-process monitoring, modeling, solving, correction, trend prediction, and hierarchical early warning of the deformation state of GIL expansion joints. The specific functions of each module are as follows: The data acquisition module, serving as the system's data source, evenly distributes multiple displacement sensors around the circumference of the left and right flanges of the target expansion joint. These sensors collect real-time data on the relative distance between corresponding measuring points on both flanges and transmit the raw sensor data to the data preprocessing module. The data preprocessing module receives the raw sensor data, performs basic signal processing, including outlier removal (data cleaning), data format standardization, and uses methods such as low-pass filtering or wavelet transform to suppress environmental noise, obtains a noise-reduced multi-point distance data sequence, and transmits the processing results to the deformation state analysis module.

[0055] The deformation state analysis module receives the preprocessed distance data, establishes a three-dimensional relative coordinate system with the center of the left flange as the origin, and sets the initial pose by combining the sensor geometry and expansion joint structural parameters. Based on the rigid body kinematics principle, it constructs a spatial deformation motion model with 6 degrees of freedom (3 translational components and 3 rotational components) and establishes an objective function with pose parameters as variables. Then, using the deformation state of the previous moment as the initial value, it solves the objective function through a nonlinear least squares optimization algorithm, outputs the preliminary 6-degree-of-freedom deformation state estimate of the current moment, and transmits it to the deformation state correction module.

[0056] The deformation state correction module receives the preliminary deformation state, constructs a nonlinear state-space model, and establishes state transition equations and measurement equations. It then uses an extended Kalman filter (EKF) for recursive estimation: first, it predicts the prior state at the current moment, then dynamically calculates the Kalman gain based on the measurement data, fusing the prediction and measurement to output a smooth and robust posterior deformation state estimate. The corrected deformation state data is then transmitted to the deformation trend analysis module. The deformation trend analysis module receives time-series deformation data output by the deformation state correction module, integrates a Long Short-Term Memory (LSTM) network with a deformation pattern recognition algorithm, and constructs a dynamic prediction model. This model is trained based on historical and current 6-DOF deformation sequences, effectively capturing deformation evolution patterns and predicting future deformation trends in multi-step expansion joints, including key indicators such as deformation velocity and cumulative displacement increment. The prediction results (such as future deformation values ​​and their trends) are then transmitted to the early warning module to support risk assessment and decision-making.

[0057] The early warning module receives the trend analysis results transmitted by the deformation trend analysis module and compares the predicted deformation trend with preset safety thresholds (such as the maximum allowable deformation value and the dangerous deformation rate). When the prediction result reaches or exceeds the threshold, an early warning signal is automatically triggered to provide timely reminders of abnormal deformation of the target expansion joint and ensure the safe operation of the equipment or structure.

[0058] like Figure 5 As shown, this embodiment also provides another GIL expansion joint omnidirectional deformation state analysis system, including: a data acquisition module, used to collect the relative distance data between corresponding measuring points on both sides of the target expansion joint in real time; The model building module is used to establish a three-dimensional relative coordinate system with one of the flange centers on the left and right sides of the target expansion joint as the origin. It sets the initial coordinates based on the known geometric layout of multiple measuring points on the flange, introduces 6-DOF pose parameters, and constructs the expansion joint deformation motion model based on the rigid body kinematics principle. It also constructs an objective function for the expansion joint deformation motion model with 6-DOF pose parameters as variables. The morphology solution module is used to solve the objective function of the expansion joint deformation motion model based on the relative distance data between multiple measurement points at the current time and historical time, using a nonlinear least squares optimization algorithm, so as to obtain the 6-DOF deformation state of the target expansion joint. The data correction module is used to dynamically smooth and optimize the 6-DOF deformation state of the target expansion joint in order to obtain the real-time spatial deformation state sequence of the expansion joint. The morphology prediction module is used to input the real-time spatial deformation state sequence of the expansion joint into a pre-trained spatial deformation state prediction model to output the spatial deformation state prediction result of the target expansion joint; wherein, the spatial deformation state prediction model is trained based on historical deformation time series data; The early warning module is used to generate structural health assessment reports and early warning information based on the prediction results of the 6-DOF deformation state and spatial deformation state of the target expansion joint.

[0059] The present invention also provides a device for analyzing the omnidirectional deformation state of a GIL expansion joint, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the method for analyzing the omnidirectional deformation state of a GIL expansion joint.

[0060] The present invention also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the GIL expansion joint omnidirectional deformation state analysis method.

[0061] When the processor executes the computer program, it implements the steps of the above-mentioned GIL expansion joint omnidirectional deformation state analysis, such as: real-time acquisition of relative distance data between corresponding measuring points on both sides of the target expansion joint flange; establishing a three-dimensional relative coordinate system with one of the flange centers on the left and right sides of the target expansion joint as the origin, setting initial coordinates based on the known geometric layout of multiple measuring points on the flange, introducing 6-DOF pose parameters, and constructing an expansion joint deformation motion model based on the rigid body kinematics principle; constructing an objective function for the expansion joint deformation motion model with 6-DOF pose parameters as variables; and using non-linear kinematics to perform the analysis based on the relative distance data between multiple measuring points at the current and historical times. A linear least squares optimization algorithm is used to solve the objective function of the expansion joint deformation motion model to obtain the 6-DOF deformation state of the target expansion joint. The 6-DOF deformation state of the target expansion joint is dynamically smoothed and optimized to obtain a real-time spatial deformation state sequence of the expansion joint. This real-time spatial deformation state sequence is input into a pre-trained spatial deformation state prediction model to output the spatial deformation state prediction result of the target expansion joint. The spatial deformation state prediction model is trained based on historical deformation time-series data. Based on the 6-DOF deformation state and the spatial deformation state prediction result of the target expansion joint, a structural health assessment report and early warning information are generated.

[0062] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing preset functions, the instruction segments describing the execution process of the computer program in the GIL expansion joint omnidirectional deformation state analysis device. For example, the computer program can be divided into a data acquisition module, a model building module, a morphology solving module, a data correction module, a morphology prediction module, and an early warning module; the specific functions are as follows: The data acquisition module is used to collect the relative distance data between corresponding measuring points on both sides of the target expansion joint flange in real time; the model building module is used to establish a three-dimensional relative coordinate system with one of the flange centers on the left and right sides of the target expansion joint as the origin, set the initial coordinates based on the known geometric layout of multiple measuring points on the flange, introduce 6-DOF pose parameters, and construct an expansion joint deformation motion model based on the rigid body kinematics principle; construct an objective function for the expansion joint deformation motion model with 6-DOF pose parameters as variables; the morphology solving module is used to calculate the relative distance between multiple measuring points at the current time and historical time. The relative distance data between the joints is used to solve the objective function of the expansion joint deformation motion model using a nonlinear least squares optimization algorithm to obtain the 6-DOF deformation state of the target expansion joint. A data correction module is used to dynamically smooth and optimize the 6-DOF deformation state of the target expansion joint to obtain a real-time spatial deformation state sequence. A morphology prediction module is used to input the real-time spatial deformation state sequence of the expansion joint into a pre-trained spatial deformation state prediction model to output the spatial deformation state prediction result of the target expansion joint; wherein, the spatial deformation state prediction model is trained based on historical deformation time series data. An early warning module is used to generate a structural health assessment report and early warning information based on the 6-DOF deformation state and spatial deformation state prediction result of the target expansion joint.

[0063] The GIL (Glass Inlet Joint) omnidirectional deformation state analysis device can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. The GIL omnidirectional deformation state analysis device may include, but is not limited to, processors and memory. Those skilled in the art will understand that the above are examples of GIL omnidirectional deformation state analysis devices and do not constitute a limitation on the GIL omnidirectional deformation state analysis device. It may include more components than described above, or combine certain components, or different components. For example, the GIL omnidirectional deformation state analysis device may also include input / output devices, network access devices, buses, etc.

[0064] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or any conventional processor. The processor is the control center for the omnidirectional deformation state analysis of the GIL expansion joint, connecting various parts of the entire GIL expansion joint omnidirectional deformation state analysis equipment via various interfaces and lines.

[0065] The memory can be used to store the computer program and / or modules. The processor realizes various functions of the GIL expansion joint omnidirectional deformation state analysis device by running or executing the computer program and / or modules stored in the memory and calling the data stored in the memory.

[0066] The memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function (such as sound playback, image playback, etc.). The data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital cards (SD cards), flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0067] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method for analyzing the omnidirectional deformation state of a GIL expansion joint.

[0068] If the modules / units integrated in the GIL expansion joint omnidirectional deformation state analysis system are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0069] Based on this understanding, the present invention can implement all or part of the processes in the above-described GIL expansion joint omnidirectional deformation state analysis method, or it can be accomplished by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of the above-described GIL expansion joint omnidirectional deformation state analysis method. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or a preset intermediate form, etc.

[0070] The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0071] It should be noted that the content contained in the computer-readable storage medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0072] In summary, this invention provides a method and related apparatus for analyzing the omnidirectional deformation state of GIL expansion joints, which has the following advantages compared with existing analysis and monitoring methods: First, the present invention establishes a three-dimensional relative coordinate system with the center of the left flange of the target expansion joint as the origin. Combined with the layout of the circumferential distributed displacement sensor and the 6-DOF spatial deformation mathematical model, it can simultaneously characterize omnidirectional spatial pose changes such as axial expansion, radial offset and angular misalignment, and realize comprehensive quantitative monitoring of the deformation state of the expansion joint. Secondly, while ensuring monitoring accuracy, this invention only requires a standard displacement sensor in conjunction with the algorithm framework proposed in this invention, without relying on high-cost dedicated equipment, which effectively reduces the complexity of system deployment and implementation costs. Third, this invention replaces traditional manual inspection with automated data acquisition and intelligent analysis, significantly reducing the subjectivity and random errors introduced by human operation. At the same time, it integrates an extended Kalman filter to make optimal estimates of the real-time deformation state and combines an LSTM network to dynamically predict future trends. The two work together to build a highly robust data processing system, which together ensures the reliability and stability of the system in the long term from the algorithm and system architecture levels.

[0073] The above embodiments are merely one of the implementation methods for achieving the technical solution of the present invention. The scope of protection claimed by the present invention is not limited to this embodiment, but also includes any variations, substitutions and other implementation methods that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A GIL expansion joint omnidirectional deformation state analysis method, characterized in that, include: Real-time acquisition of relative distance data between corresponding measuring points on both sides of the target expansion joint flange; A three-dimensional relative coordinate system is established with one of the flange centers on the left and right sides of the target expansion joint as the origin. The initial coordinates are set in combination with the known geometric layout of multiple measuring points on the flange. Six degrees of freedom pose parameters are introduced, and the expansion joint deformation motion model is constructed based on the rigid body kinematics principle. An objective function with six degrees of freedom pose parameters as variables is constructed for the expansion joint deformation motion model. Based on the relative distance data between multiple measurement points at the current time and historical time, a nonlinear least squares optimization algorithm is used to solve the objective function of the expansion joint deformation motion model to obtain the 6-DOF deformation state of the target expansion joint; wherein, the 6-DOF deformation state is represented by 6-DOF pose parameters. The 6-DOF deformation state of the target expansion joint is dynamically smoothed and optimized to obtain the real-time spatial deformation state sequence of the expansion joint. The real-time spatial deformation state sequence of the expansion joint is input into a pre-trained spatial deformation state prediction model to output the spatial deformation state prediction result of the target expansion joint; wherein, the spatial deformation state prediction model is trained based on historical deformation time series data; Based on the predicted results of the 6-DOF deformation state and spatial deformation state of the target expansion joint, a structural health assessment report and early warning information are generated.

2. The GIL expansion joint omnidirectional deformation state analysis method according to claim 1, characterized in that, The real-time acquisition of relative distance data between corresponding measuring points on both sides of the target expansion joint flange includes: Multiple displacement sensors are pre-arranged symmetrically on the circumference of the flanges on both sides of the target expansion joint; The relative distance data between corresponding measuring points on both sides of the flange of the target expansion joint is collected in real time by displacement sensors to obtain the original distance sequence. The original distance sequence is preprocessed by filtering, and the filtered distance sequence is output for solving the objective function of the extensor deformation motion model.

3. The GIL expansion joint omnidirectional deformation state analysis method of claim 1, wherein, A three-dimensional relative coordinate system is established with one of the flange centers on the left and right sides of the target expansion joint as the origin. Initial coordinates are set based on the known geometric layout of multiple measuring points on the flange. Six degrees of freedom pose parameters are introduced, and a deformation motion model of the expansion joint is constructed based on the principles of rigid body kinematics, including: A three-dimensional relative coordinate system is established with the center of the left flange of the target expansion joint as the origin; the construction principle of the coordinate system with the center of the right flange of the target expansion joint as the origin is the same as that of the left flange. The relative deformation state of the expansion joint is described as a 6-DOF variable: translation vector. Indicates to respectively Translation distance in direction, rotation vector Indicates to respectively and The angle of the axis, based on a given rotation vector. Constructing a rotation matrix using Rodriguez's formula : In the formula, denotes the unit matrix, denotes the rotation angle, which is used to indicate the rotation amplitude of the right flange relative to the left flange, ; denotes the anti-symmetric matrix of the rotation vector ; When deformation occurs, according to the rotation matrix the theoretical position of the right flange is calculated as and the theoretical distance between the current pose of the right flange and the left corresponding point is In the formula, This indicates the three-dimensional coordinates of the current measuring point on the right flange; Indicates the current left flange and The corresponding three-dimensional coordinates of the measuring point; Represents Euclidean distance; The objective function for constructing the deformation motion model of the extension joint based on the principles of rigid body kinematics is as follows: In the formula, This indicates the actual distance measured by the displacement sensor at each measuring point; For torsional motion, the regularization coefficient is denoted as . is the regularization coefficient for shear motion; This represents the pose parameters.

4. The method for analyzing the omnidirectional deformation state of a GIL expansion joint according to claim 1, characterized in that, Based on the relative distance data between multiple measurement points at the current and historical times, a nonlinear least squares optimization algorithm is used to solve the objective function of the expansion joint deformation motion model to obtain the 6-DOF deformation state of the target expansion joint, including: A nonlinear least squares optimization method is employed, using a vector of 6-DOF deformation parameters. To optimize the variables, parameters are iteratively adjusted to ensure the theoretical calculated distance at each measuring point is optimized. Compared with the measured distance The objective function minimizes the residuals between the two values; in each iteration, the parameters are updated based on the current residuals and the Jacobian matrix until the objective function converges to a preset threshold, resulting in a 6-DOF vector after iterative convergence. This allows for the quantitative determination of the deformation state. The converged rotation vector Transformed into a rotation matrix using Rodriguez's formula Combined with translation vector The six-degree-of-freedom deformation state of the target expansion joint is constituted; based on the six-degree-of-freedom deformation state, the current pose of the flanges on both sides of the target expansion joint is determined.

5. The method for analyzing the omnidirectional deformation state of a GIL expansion joint according to claim 1, characterized in that, The dynamic smoothing and optimization of the 6-DOF deformation state of the target expansion joint to obtain the real-time spatial deformation state sequence of the expansion joint includes: Predict the prior state at the current moment based on the posterior state at a historical moment; Using the 6-DOF deformation state of the target expansion joint as the observation input, the Kalman gain is dynamically calculated, and the prediction and observation information are fused to generate a corrected posterior state estimate. The corrected posterior state estimate is output as the real-time spatial deformation state sequence of the expansion joint to achieve dynamic smoothing and optimization of the 6-DOF deformation state of the target expansion joint.

6. The method for analyzing the omnidirectional deformation state of a GIL expansion joint according to claim 1, characterized in that, Before inputting the real-time spatial deformation state sequence of the expansion joint into the pre-trained spatial deformation state prediction model to output the spatial deformation state prediction result of the target expansion joint, the method further includes: constructing and training the spatial deformation state prediction model, the specific steps of which are as follows: Supervised learning samples are constructed using a sliding time window approach, wherein the supervised learning samples include historical deformation time series data and the corresponding deformation states at several future moments; The pre-constructed initial model for predicting spatial deformation state is trained using supervised learning samples, and the trained spatial deformation state prediction model is output; wherein, the base model of the initial model for predicting spatial deformation state adopts a long short-term memory network model.

7. The method for analyzing the omnidirectional deformation state of a GIL expansion joint according to claim 1, characterized in that, The predicted results of the 6-DOF deformation state and spatial deformation state based on the target expansion joint generate a structural health assessment report and early warning information, including: A two-dimensional, multi-level early warning standard covering both cumulative deformation and deformation rate is established, divided into three levels: Normal monitoring level: Deformation is within the allowable range, and the system records normally; Trend Deviation Level: Deformation or rate exceeds the normal threshold but does not reach a dangerous level, triggering a low-risk alarm; Critical risk level: The predicted deformation or rate is close to the structural failure limit, triggering a high-risk alarm; The predicted results of the 6-DOF deformation state and spatial deformation state of the target expansion joint are compared in real time with the deformation threshold and deformation rate threshold to generate early warning information of the corresponding level. A structural health assessment report is generated based on the predicted results of the 6-DOF deformation state and spatial deformation state of the target expansion joint.

8. A system for analyzing the omnidirectional deformation state of a GIL (Gas Inertia Joint) expansion joint, characterized in that, include: The data acquisition module is used to collect the relative distance data between corresponding measuring points on both sides of the target expansion joint in real time. The model building module is used to establish a three-dimensional relative coordinate system with one of the flange centers on the left and right sides of the target expansion joint as the origin. It sets the initial coordinates based on the known geometric layout of multiple measuring points on the flange, introduces 6-DOF pose parameters, and constructs the expansion joint deformation motion model based on the rigid body kinematics principle. It also constructs an objective function for the expansion joint deformation motion model with 6-DOF pose parameters as variables. The morphology solution module is used to solve the objective function of the expansion joint deformation motion model based on the relative distance data between multiple measurement points at the current time and historical time, using a nonlinear least squares optimization algorithm, so as to obtain the 6-DOF deformation state of the target expansion joint; wherein, the 6-DOF deformation state is represented by 6-DOF pose parameters; The data correction module is used to dynamically smooth and optimize the 6-DOF deformation state of the target expansion joint in order to obtain the real-time spatial deformation state sequence of the expansion joint. The morphology prediction module is used to input the real-time spatial deformation state sequence of the expansion joint into a pre-trained spatial deformation state prediction model to output the spatial deformation state prediction result of the target expansion joint; wherein, the spatial deformation state prediction model is trained based on historical deformation time series data; The early warning module is used to generate structural health assessment reports and early warning information based on the prediction results of the 6-DOF deformation state and spatial deformation state of the target expansion joint.

9. A device for analyzing the omnidirectional deformation state of a GIL expansion joint, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the omnidirectional deformation state analysis method for GIL expansion joints according to any one of claims 1-7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it is used to implement the omnidirectional deformation state analysis method for GIL expansion joints as described in any one of claims 1-7.

Images