A multi-parameter monitoring method and system for seismic isolation bearings

Abstract

This invention discloses a multi-parameter monitoring method and system for seismic isolation bearings. The method includes: dividing multiple seismic isolation bearings into associated node groups and specifying a reference node and reference time; calculating in real time the time difference between the slave nodes and the reference node in the physical data within the physical associated node group, determining whether there is time drift in the slave nodes (the slave nodes are nodes other than the reference nodes); marking the slave nodes with time drift as nodes to be repaired, finding the optimal alignment path between the physical data sequence of the node to be repaired and the template sequence; establishing a nonlinear time mapping function; and being able to perform full waveform correction for complex time distortions such as delay, stretching, and compression, thereby suppressing the formation of data tearing at its source.

Description

Technical Field

[0001] This invention relates to the field of building earthquake monitoring technology, specifically to a multi-parameter monitoring method and system for seismic isolation bearings. Background Technology

[0002] In the field of monitoring seismic isolation bearings of large ancient buildings, distributed monitoring systems often use wireless networks to achieve synchronization between nodes. However, due to the complex structure of ancient buildings and the harsh electromagnetic signal propagation environment, communication between nodes is susceptible to multipath effects and signal reflection interference. In dynamic events such as earthquakes, increased network latency fluctuations can cause clock drift in each node, resulting in inconsistent data timestamps and data tearing. This makes it impossible for the system to reconstruct a complete sequence of structural response events. Traditional methods often employ simple timestamp alignment or data discarding strategies, which are insufficient to solve the nonlinear time drift problem and cannot identify physical logic errors caused by sensor malfunctions, seriously affecting the reliability and accuracy of the monitoring system.

[0003] Therefore, proposing a multi-parameter monitoring method and system for seismic isolation bearings to address the difficulties in existing technologies is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-parameter monitoring method and system for seismic isolation bearings to address the shortcomings of the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a multi-parameter monitoring method for seismic isolation bearings, comprising:

[0006] Divide multiple seismic isolation bearings into associated node groups and specify the reference node and reference time;

[0007] Calculate the time difference in physical data between the slave node and the reference node within the associated node group to determine whether there is time drift in the slave node. The slave node is a node other than the reference node.

[0008] Slave nodes with time drift are marked as nodes to be repaired. Using the physical data sequence of the reference node as a template sequence, the optimal alignment path between the physical data sequence of the node to be repaired and the template sequence is found through a dynamic time warping algorithm. The physical data sequence of the node to be repaired is repaired according to the optimal alignment path to obtain the repaired physical data sequence.

[0009] Physical rules for seismic isolation bearings are established in advance. The repaired physical data is then input into the physical rules for testing to determine whether the physical rules are violated.

[0010] Retrieve the corresponding copy data of the physical data that violates the physical rules. If the physical data in the copy data is consistent with and conforms to the physical rules, then replace it with the physical data in the copy data; if it is inconsistent or violates the physical rules, then input the dependent variable data sequence in the physical data into the pre-trained parameterized constitutive model, output the dependent variable data that matches the current motion state and conforms to the physical laws, and replace the original dependent variable data.

[0011] Furthermore, based on the building structural design drawings, multiple seismic isolation bearings with rigid connections are grouped into the same physical association node group;

[0012] Within each associated node group, select a seismic isolation bearing node and designate it as the reference node, with its timestamp serving as the reference time for the associated node group.

[0013] Furthermore, the optimal alignment path finding methods include:

[0014] Construct a comprehensive feature vector for the reference node and slave nodes, including: a three-axis acceleration composite vector, a horizontal displacement vector magnitude, a vertical pressure reading, and a signal differential energy.

[0015] The comprehensive feature vector sequence of the node to be repaired is compared with the template sequence. For each time point i in the template sequence and each time point j in the comprehensive feature vector sequence of the node to be repaired, the Euclidean distance between the feature vectors of i and j is calculated. The Euclidean distance is filled into the (i, j) position of the two-dimensional matrix, which is the local cost matrix.

[0016] Find the path with the minimum cumulative cost from (1, 1) to (m, n) in the cost matrix, which is the optimal alignment path. m is the length of the template sequence and n is the length of the comprehensive feature vector sequence of the node to be repaired.

[0017] Furthermore, the triaxial acceleration synthesis vector combines the acceleration readings in the X, Y, and Z directions into a single scalar, representing the total vibration energy; the horizontal displacement vector magnitude is the vector magnitude for calculating displacements in two directions within the horizontal plane, representing the total deformation; and the signal differential energy is the first-order derivative of the triaxial acceleration synthesis vector, the horizontal displacement vector magnitude, and the vertical pressure reading over a preset time period.

[0018] Furthermore, finding the path with the minimum cumulative cost from (1, 1) to (m, n) in the cost matrix follows these rules:

[0019] The path must move monotonically over time; the path can only move to adjacent grid points; and the slope of the path must not exceed the limit.

[0020] Furthermore, methods for obtaining the repaired physical data include:

[0021] Optimal alignment path Where i is the time index of the baseline node; j is the original time index of the node to be repaired; The number of steps for the optimal alignment path between the template sequence and the comprehensive feature vector sequence of the node to be repaired; each point (i, j) represents the position where the j-th data point of the node to be repaired should be aligned to the i-th data point of the reference node;

[0022] Generate a time mapping function based on the optimal alignment path. The time mapping function represents the original timestamp of the node to be repaired. Standard timestamp to the baseline node One-to-one correspondence;

[0023] Based on the time mapping function, spline interpolation algorithm is used to recalculate the values ​​of all physical data for each moment of the node to be repaired on the standard time axis, generating the final repaired physical data.

[0024] Furthermore, the physical rules include:

[0025] Rule 1: When a seismic isolation bearing undergoes horizontal displacement, the vertical pressure undergoes a measurable corresponding change. Define a threshold for the change in vertical pressure versus horizontal displacement.

[0026] Rule 2: The direction of acceleration is consistent with the trend of horizontal displacement change;

[0027] Rule 3: The input energy and the energy dissipated by the seismic isolation bearing are matched in magnitude; the input energy is the product of vertical pressure, acceleration and horizontal displacement change; the energy dissipated by the seismic isolation bearing is the energy consumed by the seismic isolation bearing through its internal damping and plastic deformation.

[0028] Furthermore, methods for determining whether physical rules have been violated include:

[0029] The vertical pressure, acceleration, and displacement changes are fed into the physical rules for verification one by one. If any rule is violated, the corrected physical data is determined to violate the physical rules.

[0030] Furthermore, training methods for parameterized constitutive models include:

[0031] Historical monitoring data of seismic isolation bearings in a healthy state are collected, including synchronously recorded independent variable data sequences and actual dependent variable data. The independent variable data sequences include horizontal displacement sequences and acceleration sequences, and the dependent variable data includes vertical pressure; then preprocessing is performed.

[0032] The independent variable data sequence is used as input to the parameterized constitutive model, which outputs the corresponding dependent variable data predicted by the corresponding independent variable data sequence, and uses the actual dependent variable data corresponding to the independent variable data sequence as the prediction target; the training objective is to minimize the sum of prediction errors of vertical pressure corresponding to all independent variable data sequences; the parameterized constitutive model is trained until the sum of prediction errors converges and then training stops. The parameterized constitutive model is an interpretable linear regression or multinomial regression model.

[0033] A multi-parameter monitoring system for seismic isolation bearings is provided to implement the aforementioned multi-parameter monitoring method for seismic isolation bearings. The system includes:

[0034] The association module is used to divide multiple seismic isolation bearings into associated node groups and specify the reference node and reference time;

[0035] The verification module calculates the time difference in physical data between the slave node and the reference node within the physical association node group in real time, and determines whether there is time drift in the slave node. The slave node is a node other than the reference node.

[0036] The compensation module is used to mark slave nodes with time drift as nodes to be repaired. Using the physical data sequence of the reference node as a template sequence, the module uses a dynamic time warping algorithm to find the best alignment path between the physical data sequence of the node to be repaired and the template sequence. The module then repairs the physical data sequence of the node to be repaired based on the best alignment path to obtain the repaired physical data sequence.

[0037] The verification module pre-establishes physical rules for different types of seismic isolation bearings, inputs the repaired physical data into the physical rules for detection, and determines whether the physical rules are violated.

[0038] The conflict resolution module is used to retrieve the corresponding copy data of the physical data that violates the physical rules. The number of copy data is at least 2. If the physical data in the copy data is consistent and conforms to the physical rules, the physical data in the copy data is used to replace it. If they are inconsistent or violate the physical rules, the dependent variable data sequence in the physical data is input into the pre-trained parameterized constitutive model, and the dependent variable data that matches the current motion state and conforms to the physical laws is output and replaced the original dependent variable data.

[0039] The technical effects and advantages of the multi-parameter monitoring method and system for seismic isolation bearings provided by this invention are as follows:

[0040] Based on the rigid connection characteristics of building structures, physical associated node groups are divided. Using the benchmark node sequence as a template, a nonlinear time mapping function is established by employing comprehensive feature vectors and a constrained dynamic time warping algorithm. This overcomes the limitations of traditional simple time translation and can perform full waveform correction for complex time distortions such as delay, stretching, and compression, thereby suppressing the formation of data tearing at its source.

[0041] Based on the mechanical essence of seismic isolation bearings, causal constraint rules between displacement, pressure, and acceleration are constructed, including displacement-pressure correlation, acceleration-displacement causality, and energy conservation verification. This enables accurate identification of data anomalies, particularly detecting hidden faults that violate physical laws but are consistent with the data itself, which cannot be found by traditional methods. The physical rules effectively improve the accuracy of sensor hardware fault identification.

[0042] By employing triple-copy storage to mitigate transient errors and combining it with a parameterized constitutive model trained on health data, reliable dependent variable data can be reconstructed from independent variable data in the event of sensor failure. This forms a layered processing capability for transmission errors and copy repair, as well as hardware failures and model reconstruction. Even in sensor failure scenarios, it can still effectively maintain the integrity and physical consistency of earthquake monitoring data, significantly improving the system's survivability under extreme conditions. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0044] Figure 1 This is a schematic diagram of a multi-parameter monitoring system for seismic isolation bearings according to the present invention;

[0045] Figure 2 This is a flowchart of a multi-parameter monitoring method for seismic isolation bearings according to the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] It should be noted that when a component is said to be "fixed to" another component, it can be directly attached to the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component.

[0048] Example 1

[0049] like Figure 1 As shown in the figure, this embodiment provides a multi-parameter monitoring method and system for seismic isolation bearings, including an association module, a verification module, a compensation module, a validation module, and a conflict repair module, with the modules connected via wired and / or wireless means.

[0050] The association module is used to group multiple seismic isolation bearings rigidly connected by components such as main beams and trusses into the same physical association node group based on the building structural design drawings. The seismic isolation bearings within the association node group should exhibit a coordinated and consistent movement trend during an earthquake.

[0051] Then, within each associated node group, a seismic isolation bearing node with the most stable communication signal and the most central location is selected and designated as the reference node. The data stream collected by the reference node will be processed first, and its timestamp will be regarded as the reference time of the associated node group.

[0052] The verification module calculates in real time the time difference of physical data between the slave node and the reference node in the associated node group, such as the time difference in the peak acceleration and displacement direction, to determine whether the slave node has time drift. When it is determined that there is a delay in the data of a slave node, the data is not simply discarded, but the subsequent repair program is started.

[0053] The methods for determining whether the slave node has time drift include:

[0054] When the time difference between the physical data of the slave node and the acquisition time of the reference node exceeds a predetermined time threshold, it is determined that the slave node has time drift.

[0055] Within a rigid structure, the propagation of vibration waves is instantaneous. Therefore, the main acceleration peaks of all isolation bearings within the group should appear successively within an extremely short time window, such as ±10 milliseconds. For example, the reference node records the main acceleration peak at t=100ms, while the main peak of node A appears at t=125ms. This 25ms difference is structurally impossible, thus indicating that the data from node A has a severe time drift.

[0056] The compensation module marks slave nodes with time drift as nodes to be repaired. Using the physical data sequence of the reference node as a template sequence, a dynamic time warping algorithm is used to find the optimal alignment path between the physical data sequence of the node to be repaired and the template sequence. Based on the optimal alignment path, the physical data sequence of the node to be repaired is translated and stretched / compressed as a whole, aligning its acceleration peaks, displacement inflection points, and other features with the reference node on the time axis, resulting in the repaired physical data sequence; thus, the data distorted due to communication problems is repaired.

[0057] The best alignment path finding methods include:

[0058] A comprehensive feature vector is constructed for the reference node and slave nodes, including: a triaxial acceleration composite vector, a horizontal displacement vector magnitude, a vertical pressure reading, and a signal differential energy. The triaxial acceleration composite vector combines the acceleration readings in the X, Y, and Z directions into a single scalar, representing the total vibration energy. The horizontal displacement vector magnitude calculates the displacement in two directions within the horizontal plane, representing the total deformation. The signal differential energy calculates the first derivative (rate of change) of the above physical quantities within a short time window (e.g., 10 ms) to capture the dynamic trends of the signal, such as sudden increases in acceleration or inflection points in displacement.

[0059] It should be noted that alignment is not just about aligning a single signal (such as acceleration), but rather about constructing a comprehensive feature vector that can fully reflect the state of the seismic isolation bearing, in order to ensure the physical rationality and accuracy of the alignment. Using the comprehensive feature vector for normalization ensures that the final alignment result is consistent with the reference node in terms of acceleration, horizontal displacement, vertical pressure and their changing trends, rather than just waveform similarity, thus avoiding the error of misaligning the displacement and pressure relationship due to aligning only acceleration.

[0060] The comprehensive feature vector sequence of the node to be repaired is compared with the template sequence. For each time point i in the template sequence and each time point j in the comprehensive feature vector sequence of the node to be repaired, the Euclidean distance between the feature vectors of i and j is calculated. The Euclidean distance is filled into the (i, j) position of the two-dimensional matrix. The two-dimensional matrix is ​​the local cost matrix. The larger the Euclidean distance, the less similar the two points are.

[0061] Find the path with the minimum cumulative cost from (1, 1) to (m, n) in the cost matrix; this is the optimal alignment path. The optimal alignment path defines how to bend the time axis of the composite feature vector sequence of the repair node to best match the template sequence, where m is the length of the template sequence and n is the length of the composite feature vector sequence of the node to be repaired.

[0062] Finding the path with the minimum cumulative cost from (1, 1) to (m, n) in the cost matrix follows these rules:

[0063] The path must advance monotonically over time to ensure that the law of causality is not violated, that is, the result cannot occur before the cause.

[0064] The path can only move to adjacent grid points, ensuring the continuity of time deformation and avoiding jumps.

[0065] The slope of the path must not exceed the limit, which limits the maximum scaling ratio. The slope of the path is the most critical physical constraint; for example, setting the slope of the path to be between [0.5, 2] means that the time axis stretching or compression of the repaired slave node data cannot exceed twice that of the reference node or be shortened to half; this slope range is preset based on the structural dynamics principle of the maximum possible response time difference between rigidly connected seismic isolation bearings under seismic wave action.

[0066] Methods for obtaining the repaired physical data include:

[0067] Optimal alignment path Where i is the time index of the baseline node; j is the original time index of the node to be repaired; The number of steps is the optimal alignment path between the template sequence and the comprehensive feature vector sequence of the node to be repaired; each point (i, j) represents the position where the j-th data point of the node to be repaired should be aligned to the i-th data point of the reference node.

[0068] Generate a time mapping function based on the optimal alignment path. The time mapping function represents the original timestamp of the node to be repaired. Standard timestamp to the baseline node The one-to-one correspondence is expressed by the following formula:

[0069] Adjacent points on the best alignment path and , Within the interval, the mapping function The linear representation is:

[0070]

[0071] Example as follows:

[0072] Assuming there are two adjacent points on the path, (100, 150) and (101, 152), for the original time point :

[0073]

[0074] That is, the original timestamp of the node to be repaired, the 151st data point, has been corrected to time 100.5 on the reference time axis.

[0075] Based on the time mapping function, spline interpolation algorithm is used to recalculate the values ​​of all physical data for each moment of the node to be repaired on the standard time axis, generating the final repaired physical data.

[0076] The verification module pre-establishes physical rules for different types of seismic isolation bearings. The repaired physical data is input into these rules for testing, determining whether any violations occur. The physical rules are rigid and based on classical mechanics, and include:

[0077] Rule 1: When a seismic isolation bearing undergoes horizontal displacement, its vertical pressure must undergo a measurable corresponding change. Define a reasonable threshold for the change in vertical pressure versus horizontal displacement.

[0078] Rule 2: Acceleration is the second derivative of displacement, and the direction of acceleration must be consistent with the trend of horizontal displacement change.

[0079] Rule 3: The input energy and the energy dissipated by the seismic isolation bearing must match in magnitude. The input energy is the product of vertical pressure, acceleration, and horizontal displacement. The energy dissipated by the seismic isolation bearing refers to the energy consumed by the bearing through its internal damping and plastic deformation, in the form of heat. This is the core function of the seismic isolation bearing: to dissipate seismic energy and protect the superstructure. This energy can be estimated using the area of ​​the hysteresis curve, such as by numerical integration of the closed loop formed by vertical pressure and horizontal displacement (e.g., the trapezoidal method).

[0080] Methods for determining whether physical rules have been violated include:

[0081] The vertical pressure, acceleration, and displacement changes are fed into the physical rules for verification one by one. If any rule is violated, the corrected physical data is determined to violate the physical rules.

[0082] The conflict resolution module is used to retrieve the corresponding copy data of the physical data that violates the physical rules. The number of copy data is at least 2. If the physical data in the copy data is consistent and conforms to the physical rules, the physical data in the copy data is used to replace it. If they are inconsistent or violate the physical rules, it indicates that there may be a sensor hardware failure. In this case, the dependent variable data sequence in the physical data is input into the pre-trained parameterized constitutive model, and the theoretical value of the dependent variable data that matches the current motion state and conforms to the physical laws is output and replaced with the original dependent variable data for subsequent safety assessment.

[0083] Training methods for parameterized constitutive models include:

[0084] Historical monitoring data of seismic isolation bearings in a healthy state are collected, including synchronously recorded independent variable data sequences and actual dependent variable data. The independent variable data sequences include horizontal displacement sequences and acceleration sequences, and the dependent variable data includes vertical pressure; then preprocessing is performed.

[0085] The independent variable data sequence is used as input to the parameterized constitutive model, which outputs the corresponding dependent variable data predicted by the corresponding independent variable data sequence, and uses the actual dependent variable data corresponding to the independent variable data sequence as the prediction target; the training objective is to minimize the sum of prediction errors of vertical pressure corresponding to all independent variable data sequences; the parameterized constitutive model is trained until the sum of prediction errors converges and then training stops. The parameterized constitutive model is an interpretable linear regression or multinomial regression model.

[0086] It should be noted that in the multi-parameter monitoring system of seismic isolation bearings, horizontal displacement and acceleration are defined as independent variable data, which characterize the motion state of the structure; while vertical pressure is defined as dependent variable data, and its value is the instantaneous response generated by mapping the independent variable data through the mechanical constitutive relation of the seismic isolation bearing.

[0087] When the physical rule verification finds a logical conflict between the dependent variable data and the independent variable data, the system corrects the vertical pressure data to restore the physical consistency of the outlier, thereby fundamentally ensuring the physical authenticity of the data chain and the reliability of the system evaluation. This avoids systemic misjudgment caused by incorrect repair of motion state parameters. This principle ensures that the system does not introduce new physical contradictions while repairing the data.

[0088] The replica data is stored in the node's local memory and on a separate secure storage chip on the node, so that a single hardware failure will not simultaneously corrupt other data. When a sensor (such as a displacement meter) collects a reading, this data packet is accompanied by a timestamp and then immediately sent to the monitoring system, where two identical copies are made and stored independently.

[0089] Example 2

[0090] Please see Figure 2 As shown, this embodiment provides a multi-parameter monitoring method for seismic isolation bearings. Details not shown are described in Embodiment 1. The method includes:

[0091] Divide multiple seismic isolation bearings into associated node groups and specify the reference node and reference time;

[0092] Real-time calculation of the time difference in physical data between slave nodes and the reference node within a physically associated node group; determination of whether slave nodes have time drift; slave nodes are nodes other than the reference node.

[0093] Slave nodes with time drift are marked as nodes to be repaired. Using the physical data sequence of the reference node as a template sequence, the optimal alignment path between the physical data sequence of the node to be repaired and the template sequence is found through a dynamic time warping algorithm. The physical data sequence of the node to be repaired is repaired according to the optimal alignment path to obtain the repaired physical data sequence.

[0094] Physical rules are pre-established for different types of seismic isolation bearings. The repaired physical data is then input into the physical rules for testing to determine whether the physical rules are violated.

[0095] Retrieve the corresponding copy data of the physical data that violates the physical rules. The number of copy data is at least 2. If the physical data in the copy data is consistent and conforms to the physical rules, then the physical data in the copy data is used to replace it. If they are inconsistent or violate the physical rules, then the dependent variable data sequence in the physical data is input into the pre-trained parameterized constitutive model, and the dependent variable data that matches the current motion state and conforms to the physical laws is output and replaced the original dependent variable data.

[0096] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for multi-parameter monitoring of seismic isolation bearings, characterized in that, include: Divide multiple seismic isolation bearings into associated node groups and specify the reference node and reference time; Calculate the time difference in physical data between the slave node and the reference node within the associated node group to determine whether there is time drift in the slave node. The slave node is a node other than the reference node. Slave nodes with time drift are marked as nodes to be repaired. Using the physical data sequence of the reference node as the template sequence, the optimal alignment path between the physical data sequence of the node to be repaired and the template sequence is found through a dynamic time warping algorithm. The physical data sequence of the node to be repaired is repaired according to the optimal alignment path to obtain the repaired physical data sequence; Physical rules for seismic isolation bearings are established in advance. The repaired physical data is then input into the physical rules for testing to determine whether the physical rules are violated. Retrieve at least two copies of the physical data corresponding to the physical data that violates the physical rules. If the physical data in the copies is consistent with and conforms to the physical rules, then the physical data in the copies is used to replace the original physical data. If they are inconsistent or violate the physical rules, then the dependent variable data sequence in the physical data is input into the pre-trained parameterized constitutive model, and the output dependent variable data that matches the current motion state and conforms to the physical laws is used to replace the original dependent variable data. The training method of the parameterized constitutive model includes: Historical monitoring data of seismic isolation bearings in a healthy state are collected, including synchronously recorded independent variable data sequences and actual dependent variable data. The independent variable data sequences include horizontal displacement sequences and acceleration sequences, and the dependent variable data includes vertical pressure; then preprocessing is performed. The independent variable data sequence is used as the input of the parameterized constitutive model, which outputs the corresponding dependent variable data predicted by the corresponding independent variable data sequence, and uses the actual dependent variable data corresponding to the independent variable data sequence as the prediction target. The training objective is to minimize the sum of prediction errors of vertical pressure corresponding to all independent variable data sequences. The parameterized constitutive model is trained until the sum of prediction errors converges, at which point training stops. The parameterized constitutive model is an interpretable linear regression or multinomial regression model.

2. The multi-parameter monitoring method for seismic isolation bearings according to claim 1, characterized in that, According to the building structural design drawings, multiple seismic isolation bearings with rigid connections are assigned to the same physical associated node group; Within each associated node group, select a seismic isolation bearing node and designate it as the reference node, with its timestamp serving as the reference time for the associated node group.

3. The multi-parameter monitoring method for seismic isolation bearings according to claim 2, characterized in that, The best alignment path finding methods include: Construct a comprehensive feature vector for the reference node and slave nodes, including: a three-axis acceleration composite vector, a horizontal displacement vector magnitude, a vertical pressure reading, and a signal differential energy. The comprehensive feature vector sequence of the node to be repaired is compared with the template sequence. For each time point i in the template sequence and each time point j in the comprehensive feature vector sequence of the node to be repaired, the Euclidean distance between the feature vectors of i and j is calculated. The Euclidean distance is filled into the (i, j) position of the two-dimensional matrix, which is the local cost matrix. Find the path with the minimum cumulative cost from (1, 1) to (m, n) in the cost matrix, which is the optimal alignment path. m is the length of the template sequence and n is the length of the comprehensive feature vector sequence of the node to be repaired.

4. The multi-parameter monitoring method for seismic isolation bearings according to claim 3, characterized in that, The triaxial acceleration synthesis vector combines the acceleration readings in the X, Y, and Z directions into a single scalar, representing the total vibration energy; the horizontal displacement vector magnitude is the vector magnitude for calculating displacements in two directions within the horizontal plane, representing the total deformation; the signal differential energy is the first-order derivative of the triaxial acceleration synthesis vector, the horizontal displacement vector magnitude, and the vertical pressure reading over a preset time period.

5. The multi-parameter monitoring method for seismic isolation bearings according to claim 3, characterized in that, Finding the path with the minimum cumulative cost from (1, 1) to (m, n) in the cost matrix follows these rules: The path must move monotonically over time; the path can only move to adjacent grid points; and the slope of the path must not exceed the limit.

6. A multi-parameter monitoring method for seismic isolation bearings according to claim 3 or 5, characterized in that, Methods for obtaining the repaired physical data include: Optimal alignment path Where i is the time index of the baseline node; j is the original time index of the node to be repaired; The number of steps for the optimal alignment path between the template sequence and the comprehensive feature vector sequence of the node to be repaired; each point (i, j) represents the position where the j-th data point of the node to be repaired should be aligned to the i-th data point of the reference node; Generate a time mapping function based on the optimal alignment path. The time mapping function represents the original timestamp of the node to be repaired. Standard timestamp to the baseline node One-to-one correspondence; Based on the time mapping function, spline interpolation algorithm is used to recalculate the values ​​of all physical data for each moment of the node to be repaired on the standard time axis, generating the final repaired physical data.

7. The multi-parameter monitoring method for seismic isolation bearings according to claim 1, characterized in that, The physical rules include: Rule 1: When a seismic isolation bearing undergoes horizontal displacement, the vertical pressure undergoes a measurable corresponding change. Define a threshold for the change in vertical pressure versus horizontal displacement. Rule 2: The direction of acceleration is consistent with the trend of horizontal displacement change; Rule 3: The input energy and the energy dissipated by the seismic isolation bearing are matched in magnitude; the input energy is the product of vertical pressure, acceleration and horizontal displacement change; the energy dissipated by the seismic isolation bearing is the energy consumed by the seismic isolation bearing through its internal damping and plastic deformation.

8. The multi-parameter monitoring method for seismic isolation bearings according to claim 1, characterized in that, Methods for determining whether physical rules have been violated include: The vertical pressure, acceleration, and displacement changes are fed into the physical rules for verification one by one. If any rule is violated, the corrected physical data is determined to violate the physical rules.

9. A multi-parameter monitoring system for seismic isolation bearings, characterized in that, For implementing the multi-parameter monitoring method for seismic isolation bearings according to any one of claims 1-8, the system comprises: The association module is used to divide multiple seismic isolation bearings into associated node groups and specify the reference node and reference time; The verification module calculates the time difference in physical data between the slave node and the reference node within the physical associated node group, and determines whether there is time drift in the slave node. The slave node is a node other than the reference node. The compensation module is used to mark slave nodes with time drift as nodes to be repaired. Using the physical data sequence of the reference node as a template sequence, the module uses a dynamic time warping algorithm to find the best alignment path between the physical data sequence of the node to be repaired and the template sequence. The module then repairs the physical data sequence of the node to be repaired based on the best alignment path to obtain the repaired physical data sequence. The verification module pre-establishes physical rules for different types of seismic isolation bearings, inputs the repaired physical data into the physical rules for detection, and determines whether the physical rules are violated. The conflict resolution module is used to retrieve the corresponding copy data of the physical data that violates the physical rules. If the physical data in the copy data is consistent and conforms to the physical rules, the physical data in the copy data is used to replace it. If they are inconsistent or violate the physical rules, the dependent variable data sequence in the physical data is input into the pre-trained parameterized constitutive model, and the dependent variable data that matches the current motion state and conforms to the physical laws is output and replaced with the original dependent variable data.

Images