Structural state population identification apparatus and method

Abstract

The application discloses a kind of structural state group identification device and method.The device includes multiple force sensing components and processing units, force sensing components are arranged at each load-bearing node of the structure to be monitored, and the load and / or torque borne by each load-bearing node is sensed;The processing unit determines the change amount of the force value distribution consistency state relative to the reference state based on the real-time force value distribution and the reference force value distribution, and identifies the state change type of the structure based on the change amount and the change characteristics of each force value relative to the reference force value.The application realizes the automatic identification of overall state change and local anomaly through the change amount of the force value distribution consistency state, and is suitable for health monitoring and safety diagnosis of structures with multiple load-bearing nodes, such as bridge bearings, building structures, wind power generation equipment, intelligent warehousing, antenna support structures, etc.

Description

Technical Field

[0001] This invention belongs to the field of equipment health monitoring and structural safety diagnosis technology, specifically relating to a device and method for real-time monitoring of structural stress state using multiple force sensing components, applicable to load-bearing structures with multiple load-bearing nodes. Background Technology

[0002] In industrial equipment and engineering structures, many critical structures rely on multiple load-bearing nodes to jointly bear loads. There is an inherent mechanical relationship between the force values ​​of each load-bearing node, and the stability of the structure depends on the balanced distribution of force values ​​across these nodes. During long-term service, factors such as uneven foundation settlement, material creep, vibration relaxation, and localized damage can alter the load distribution across the load-bearing nodes. When the force on a particular load-bearing node abnormally decreases, its load is redistributed to adjacent nodes. When overall material creep or temperature changes occur, the force values ​​of all load-bearing nodes decrease synchronously, but the force distribution remains essentially unchanged. These two different types of state changes have drastically different impacts on structural safety. The former requires precise location and emergency response, while the latter only requires continuous monitoring and periodic maintenance. Failure to distinguish between them can lead to false alarms, missed alarms, or inappropriate maintenance strategies, posing significant safety hazards.

[0003] How to effectively distinguish between the two types of state changes by analyzing the changes in the spatial distribution characteristics of force values ​​is a problem that needs to be solved in this field. Summary of the Invention

[0004] To address the above problems, the present invention provides a structural state group identification device and method.

[0005] Firstly, a structural state group identification device is provided, comprising:

[0006] Multiple force sensing components are arranged at each load-bearing node of the structure to be monitored to sense the load and / or moment borne by each load-bearing node. The force sensing components refer to devices capable of directly or indirectly sensing the force or load borne by the object under test and converting the force value into a recognizable electrical signal. Their implementation forms include, but are not limited to, strain sensing, piezoelectric sensing, capacitive sensing, or optical sensing. Their deployment methods include, but are not limited to: being arranged between the load-bearing node and the supporting surface to sense axial load; being side-mounted on the side of the structure to sense shear force or overturning moment; and being suspended at the suspension point to sense axial tensile force.

[0007] The processing unit, communicatively connected to each of the force sensing components, is configured to: determine the change in the force distribution consistency state relative to the reference state based on the real-time force distribution of each force sensing component and the reference force distribution of the structure; and identify the type of state change of the structure based on the change and the change characteristics of each force relative to the reference force. Here, a global state change refers to a state where each force changes synchronously relative to the reference force, and the change does not exceed a preset change threshold; a local anomaly refers to a state where there are abnormal measurement points with force deviations exceeding a preset deviation threshold, and the change exceeds the preset change threshold. The change can be characterized by a force distribution consistency index, which is calculated in ways including but not limited to force variance, force standard deviation, force coefficient of variation, or force interquartile range. This invention captures changes in the spatial distribution pattern of forces caused by load redistribution by monitoring the change in the force distribution consistency state.

[0008] Furthermore, the change is determined by: calculating the real-time force distribution consistency index based on the real-time force value data of each load-bearing node, and calculating the benchmark force distribution consistency index based on the benchmark force value data of each load-bearing node; the ratio or difference between the two is taken as the change.

[0009] Furthermore, the synchronous change refers to the consistent trend of change of each force value and the satisfaction of a preset ratio condition. The preset ratio condition includes: the difference in the change amplitude of each force value is within a preset range, or each force value changes according to a fixed ratio.

[0010] Furthermore, the processing unit is configured to: when there is an abnormal measuring point where the force value deviation exceeds the preset deviation threshold, but the change amount does not exceed the preset change threshold, mark the measuring point as an observation point and increase the data acquisition frequency; within a preset observation window, if the change amount exceeds the preset change threshold, escalate to a local anomaly alarm; otherwise, deactivate the observation state. With the introduction of the observation state mechanism, the processing unit can begin continuous tracking when the anomaly is just beginning and not yet sufficient to disrupt the overall force value distribution pattern.

[0011] Furthermore, the processing unit is configured to: when a local anomaly is determined, analyze the coupling relationship of force changes between the anomalous measuring point and its associated measuring points based on the spatial location of each load-bearing node and a pre-constructed load transfer relationship model. When the force value of the anomalous measuring point decreases and the force value of the associated measuring point increases, and the magnitude of the force change satisfies the transfer ratio in the load transfer relationship model, the anomalous measuring point is identified as the root cause node. By identifying the root cause node, it is possible to distinguish between root-cause anomalies and dependent changes.

[0012] Furthermore, the processing unit is also configured to: extract historical change trends based on the force value time series of each force sensing component within a preset historical period, estimate the expected time when the force value deviation reaches a preset warning threshold, and output estimated information including the expected time and maintenance suggestions.

[0013] Furthermore, the processing unit is also configured to: acquire environmental parameters; and suppress the warning output when the change amount does not exceed the preset change threshold and the correlation between the force value change and the environmental parameter change exceeds a preset correlation threshold. When it is determined to be an overall state change, it distinguishes between changes caused by long-term normal aging of the structure and changes caused by occasional environmental factors; if caused by long-term normal aging of the structure, it adjusts the baseline force value distribution towards the real-time force value distribution at a preset update rate; if caused by occasional environmental factors, it keeps the baseline force value distribution unchanged.

[0014] Furthermore, the processing unit is also configured to: extract the direction, magnitude and rate of change of the force value at each measuring point relative to the reference force value distribution, form a force value change feature vector, match it with a preset abnormal pattern library, and determine the cause type of the abnormality based on the matching result.

[0015] Furthermore, the processing unit is also configured to: calculate the dominant ratio obtained by decomposing the covariance matrix eigenvalues ​​of the force values ​​at each measuring point based on the force value distribution, and use it as an indicator of the structural concentration of the force value distribution, and combine it with the change amount to conduct a multidimensional stability assessment.

[0016] Furthermore, the processing unit is also configured to: identify the periodic operating characteristics of the structure, determine the self-diagnosis window period, and perform precise diagnosis based on the force distribution obtained under standard operating conditions within the self-diagnosis window period.

[0017] Secondly, a method for identifying structural state groups is provided, comprising: under a baseline healthy state, acquiring the baseline force value distribution at each load-bearing node and calculating a baseline force value distribution consistency index; during operation, acquiring the real-time force value distribution at each load-bearing node and calculating a real-time force value distribution consistency index; determining the change in the force value distribution consistency state based on both; and identifying the state change type of the structure based on the change and the change characteristics of each force value relative to the baseline force value.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] (i) By monitoring the change in the uniformity of force distribution, it is possible to accurately capture the change in the spatial distribution of force caused by load redistribution, breaking through the limitation of traditional schemes that only focus on the force change at a single measuring point.

[0020] (II) Automatic identification of overall state changes and local anomalies. When all load-bearing nodes change synchronously due to global factors, the changes are basically stable and there will be no false alarms; when an anomaly occurs in an individual node, the changes change significantly, which can be detected sensitively. An observation-state mechanism is introduced to effectively capture early anomaly signals. Through the load transfer relationship model, the root cause node of the anomaly can be accurately located.

[0021] (III) Anomaly Cause Identification. By matching the force value change feature vector with a preset anomaly pattern library, the cause type of anomaly can be identified, providing precise guidance for maintenance decisions.

[0022] (iv) Trend prediction. By extracting the historical trend of force value changes, the expected time when the force value deviation reaches the warning threshold is predicted, providing a forward-looking quantitative basis for maintenance decisions.

[0023] (v) Normal aging identification and benchmark update. When it is determined to be an overall change in condition, the cause of the change is further distinguished. The benchmark force value distribution is only adjusted when it is determined to be caused by normal aging, so as to avoid false alarms caused by long-term cumulative deviation.

[0024] (vi) Environmental parameter decoupling. Environmental parameters are introduced as an auxiliary judgment basis. When the force value change is highly correlated with the environmental parameter change and the change is stable, the warning output is suppressed, and false alarms are significantly reduced.

[0025] (vii) Wide range of applicable scenarios and flexible deployment methods. Applicable to various structures with multiple load-bearing nodes, such as bridge bearings, building structures, wind power generation equipment, smart warehousing, and antenna support structures. The force sensing components support multiple deployment methods, including bottom mounting, side mounting, and hoisting. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the force sensing component arrangement in a multi-point support structure according to an embodiment of the present invention.

[0027] Figure 2 This is a schematic diagram of the monitoring and early warning logic flow in an embodiment of the present invention.

[0028] Figure 3 This is a schematic diagram illustrating the correspondence between the force value change trend and the warning level in an embodiment of the present invention.

[0029] The markings in the attached diagram are explained as follows:

[0030] 1—Supported structure; 2—Supporting foundation; 3—Bearing node; 4—Force sensing component; 5—Processing unit; 6—Signal acquisition cable; 7—Communication interface.

[0031] It should be understood that the accompanying drawings are for illustrative purposes only and are not intended to limit the scope of protection of the present invention. The same reference numerals in the drawings are used to refer to the same parts. Detailed Implementation

[0032] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0033] Example 1: Status Monitoring of Multi-Point Support Structures

[0034] See Figure 1 The base of a certain device is supported on a support foundation (2) by eight load-bearing nodes (3), and the eight load-bearing nodes are evenly distributed along the circumference of the bottom of the supported structure (1). Force sensing components (4) are installed between each load-bearing node and the support foundation to sense the axial load borne by each load-bearing node.

[0035] After installation and debugging, the processing unit (5) collects the readings of all force sensing components through the signal acquisition cable (6) as the reference force value distribution of the support structure and stores it in non-volatile memory. The expression of the reference force value distribution is F0=[f1(0), f2(0), ..., f8(0)]. At the same time, the consistency index of the reference force value distribution is calculated based on the reference force value data of each load-bearing node. Preferably, for example, the reference variance σ²(0)=(1 / N)∑(f_i(0)-f̄(0))², where f̄(0) is the mean of each force value in the reference force value distribution.

[0036] During equipment operation, the processing unit collects the current force values ​​of each force sensing component according to a preset sampling period. It calculates the force value deviation at each measuring point and calculates the real-time force value distribution consistency index σ²(t) based on the real-time force value data of each load-bearing node. The ratio of the current index to the benchmark index, R=σ²(t) / σ²(0), is used as the change in the force value distribution consistency state relative to the benchmark state.

[0037] The monitoring and early warning logic flow of this embodiment can be found in [reference]. Figure 2 .

[0038] The processing unit uses the following criteria for state identification:

[0039] Overall state change: When the deviation signs of all force values ​​are consistent, that is, the changing trends of all force values ​​are consistent, and the changes of all force values ​​meet the preset proportional conditions: the difference in the magnitude of the changes of each force value is within the preset range, or the force values ​​change according to a fixed proportion; at the same time, the amount of change R is within the preset change threshold range, preferably, for example, between 0.8 and 1.25; and the changes in force values ​​are not related to changes in environmental parameters, it is determined to be an overall state change. At this time, the force values ​​of each load-bearing node change synchronously, but the relative magnitude relationship between the force values ​​remains basically unchanged, and the spatial pattern of force value distribution is not disrupted.

[0040] Local anomaly: When the force deviation Δf_j(t) at a certain measuring point j exceeds a preset deviation threshold, and the change R exceeds a preset change threshold range, preferably, for example, greater than 1.25 or less than 0.8, while the force deviations of other measuring points are all within the normal range, it is determined that a local anomaly has occurred at measuring point j. The processing unit calculates the deviation of each force value from the current group mean, and the measuring point with the largest deviation is the abnormal measuring point.

[0041] When the conditions of the above two criteria are met simultaneously, the processing unit prioritizes judging the situation where individual force value deviations exceed the threshold and the change amount changes significantly as a local anomaly.

[0042] Observation-state processing: When individual force value deviations exceed a preset deviation threshold, but the change R remains within the preset change threshold range, the processing unit does not immediately determine it as a local anomaly. Instead, it marks the measuring point as an observation point and increases the sampling frequency from the normal sampling cycle to a high-frequency sampling mode, preferably, for example, from once every 10 minutes to once every 1 minute, continuously tracking the force value change trend of the measuring point within a preset observation window. If, within the preset observation window, preferably, for example, within 30 minutes, the force value deviation continues to increase and the change R begins to exceed the preset change threshold range, then a local anomaly alarm is issued; if the force value deviation recovers on its own or remains stable, the observation state is lifted, and the normal sampling frequency is restored.

[0043] Environmental parameter decoupling: The processing unit also acquires environmental parameters. When the correlation between the force value change and the environmental parameter change exceeds a preset correlation threshold, preferably, for example, a Pearson correlation coefficient greater than 0.8, and the change R is within the preset change threshold range, it is determined to be a normal phenomenon caused by environmental parameter fluctuations, the warning output is suppressed, and only the event log is recorded.

[0044] Normal Aging Identification and Baseline Update: When the processing unit repeatedly determines that there is an overall state change, preferably, for example, five times consecutively, the force value change characteristics are further analyzed to distinguish the cause of the change. If the force value change shows the characteristics of long-term continuous, slow change and gradually decreasing rate of change, it is determined to be caused by long-term normal aging of the structure; if the force value change shows the characteristics of short-term sudden change or oscillation, it is determined to be caused by occasional environmental factors. When it is determined to be caused by long-term normal aging of the structure, the processing unit adjusts the baseline force value distribution at a preset update rate. One specific adjustment method is: F0_new=α·F0_old+(1-α)·F(t), where α is the forgetting factor, preferably, the value range is, for example, 0.95 to 0.995; when it is determined to be caused by occasional environmental factors, the baseline force value distribution is kept unchanged, and subsequent observations are awaited.

[0045] Example 2: Determination of the root cause node of local anomalies

[0046] This embodiment demonstrates how to determine the root cause node of a local anomaly based on a load transfer relationship model.

[0047] A load transfer relationship model is pre-constructed based on the geometric connections and mechanical transfer paths of the structure. This model represents the load transfer direction, transfer ratio, and transfer conditions between each load-bearing node. It can be constructed based on a finite element digital twin model: in the digital twin model, the support stiffness of each node is decreased sequentially, and the force value changes of the remaining nodes are recorded. If the change amplitude exceeds a preset threshold, a transfer relationship is established, and the normalized change amplitude is used as the transfer ratio.

[0048] When the processing unit determines that a certain measuring point is an abnormal measuring point, it further analyzes the coupling relationship of force value changes between the abnormal measuring point and its associated measuring points: if the force value of the abnormal measuring point decreases while the force value of the associated measuring point increases, and the direction of change is opposite, and the total increase and decrease satisfy the transfer ratio relationship in the load transfer relationship model, then it is determined to be a subordinate change caused by load redistribution, and the measuring point with the decreasing force value is identified as the root node; if the force value of the abnormal measuring point decreases while the force value of the associated measuring point also decreases in the same direction, and the magnitude of the force value change of each measuring point is similar to the ratio of the reference force value, then it is determined to be overall settlement or creep.

[0049] Example 3: Trend Forecasting

[0050] During long-term operation, the processing unit extracts the historical trend of force value at each measuring point based on the force value time series of each force sensing component within a preset historical period. One specific implementation method is to use exponential smoothing or polynomial fitting to extrapolate and calculate the expected time when the force value deviation reaches the preset warning threshold.

[0051] For a diagram illustrating the relationship between force value change trends and warning levels, please refer to [link / reference needed]. Figure 3 .

[0052] Taking the force attenuation scenario as an example, the attenuation rate is defined as r_i=(f_i(0)-f_i(t)) / (t×f_i(0)). When the attenuation rate of a certain load-bearing node continuously exceeds the preset attenuation warning threshold and continuously exceeds the preset observation period, it is determined that the load-bearing node or support surface has entered the accelerated deterioration period. The estimated remaining safe operating time is t_rem=(Th_fail-Δf_i(t)) / (r_i×f_i(0)), where Th_fail is the preset failure force deviation threshold. The processing unit outputs the estimated information including abnormal location, current force status, estimated remaining safe operating time and maintenance suggestions through the communication interface (7).

[0053] Example 4: Monitoring of Ring-shaped Support Structure

[0054] A silo's annular supported structure is supported on a ring-shaped foundation by multiple load-bearing nodes evenly distributed along its circumference. Each load-bearing node is divided into multiple sectors according to its circumferential angle, and each sector contains several load-bearing nodes. The processing unit performs group analysis by sector: when the force values ​​of multiple load-bearing nodes within a sector decrease synchronously, and the difference between the sector's average force value and the global average force value exceeds a preset sector deviation threshold, it is determined that a regional anomaly exists in that sector. In the above sector analysis, the change in the consistency of force distribution is still calculated based on the global force distribution; the comparison between the sector average and the global average force value is used to further locate the region of the anomaly after determining a local anomaly.

[0055] Example 5: Monitoring of Uneven Settlement of Equipment Base

[0056] The base of a certain unit is fixed to a foundation platform by 12 load-bearing nodes arranged in a rectangular pattern. During long-term operation, if uneven settlement occurs in a corner of the foundation platform, the force value of the load-bearing nodes in that area will decrease significantly, while the force value of the load-bearing nodes in the opposite corner will increase, resulting in a significant change in the magnitude of the change. The processing unit can automatically identify abnormal measuring points and determine them as regional anomalies.

[0057] Example 6: Multi-point support structure and anti-tipping monitoring of shelving

[0058] A smart warehouse rack consists of several uprights and multiple shelves, preferably, for example, four uprights, each with a support leg at its base. A force sensing component is installed between each support leg and the ground. During initial installation, the force values ​​of each support leg are collected as a baseline force distribution under no-load conditions. When the force value of a particular support leg continuously decreases while the force value of the opposite support leg increases accordingly, a significant change in the change is observed. Based on this, the processing unit determines that the center of gravity has shifted, posing a risk of tipping over, and can trigger corresponding alarms or protective actions.

[0059] Example 7: Side mounting and suspension methods for force sensing components

[0060] To monitor the horizontal slippage trend of the equipment base or the lateral overturning moment of the structure, the force sensing component can be mounted sideways on the side of the structure, with the sensitive axis parallel to the horizontal direction, to sense shear force or horizontal thrust. For suspended equipment or pipe hangers, the force sensing component can be suspended at the suspension point, connected in series between the suspended object and the suspension structure, to sense axial tensile force. The processing unit automatically identifies suspension point loosening and overall load changes by monitoring changes in the consistency of force distribution at each suspension point.

[0061] Example 8: Multidimensional Stability Assessment

[0062] This embodiment, based on Embodiment 1, further introduces a force distribution structure concentration index for multidimensional stability assessment. The processing unit periodically calculates the covariance matrix of the force values ​​at each measurement point and performs eigenvalue decomposition to calculate the dominant ratio, i.e., the ratio of the largest eigenvalue to the sum of all eigenvalues, as the force distribution structure concentration index. The closer the dominant ratio is to 1, the more concentrated and stable the force distribution structure; a significant change in the dominant ratio indicates a significant change in the force distribution structure. The processing unit combines the change in R with the change in the dominant ratio: if both R and the change in the dominant ratio do not exceed the threshold and all force values ​​change synchronously, it is determined to be a change in the overall state; if R exceeds the threshold and the dominant ratio also changes significantly, it is determined to be a local anomaly with high reliability; if only R slightly exceeds the threshold but the dominant ratio does not change significantly, it may be a transient disturbance, and the processing unit will not trigger a response, but only record and continuously observe.

[0063] Example 9: Anomaly Pattern Matching and Cause Identification

[0064] This embodiment demonstrates how the processing unit identifies the specific causes of anomalies through anomaly pattern matching. The processing unit extracts the direction, amplitude, and rate of force change at each measuring point relative to the baseline force distribution, forming a force change feature vector, which is then matched against a preset anomaly pattern library. The preset anomaly pattern library stores multiple force distribution change feature patterns corresponding to different cause types, and the matching process can employ similarity measurement methods such as cosine similarity or Euclidean distance. The examples of the anomaly pattern library are as follows: Local foundation erosion mode: force values ​​at individual measuring points drop sharply, while force values ​​at opposite measuring points increase, resulting in a significant change in the magnitude of change; Overall uniform settlement mode: force values ​​at all measuring points decrease synchronously with similar magnitudes, resulting in a relatively constant magnitude of change; Material creep mode: force values ​​at all measuring points decrease slowly at a gradually decreasing rate, resulting in a relatively constant magnitude of change; Connection loosening mode: force values ​​at individual measuring points decrease continuously and slowly, while force values ​​at adjacent measuring points increase accordingly, resulting in a significant change in the magnitude of change; Overload mode: force values ​​at all measuring points increase synchronously, resulting in a relatively constant magnitude of change; Temperature fluctuation mode: force values ​​at all measuring points change synchronously and periodically, highly correlated with temperature changes, resulting in a relatively constant magnitude of change. Based on the matching results, the processing unit outputs the cause type of the anomaly and its confidence level.

[0065] Example 10: Combined Application of Multiple Monitoring Modes

[0066] In practical engineering, a complex structure may simultaneously contain multiple stress forms such as multi-point support, lateral fixation, and suspension points. The health monitoring device of this invention can uniformly connect force sensing components with various deployment methods to a single processing unit, establishing a global baseline force value distribution covering all stress points for comprehensive health assessment. The load transfer relationship model can cover the force transmission relationships between different types of load-bearing nodes, enabling anomaly tracing across monitoring modes.

[0067] Example 11: Adjustability of Parameters

[0068] The numerical parameters mentioned in this specification are recommended values ​​based on typical industrial scenarios. Those skilled in the art can adjust these parameters within a reasonable range according to factors such as the specifications, materials, load characteristics, and safety requirements of the specific application. Such adjustments do not depart from the spirit and scope of protection of this invention.

[0069] It should be noted that although the above embodiments are illustrated using industrial equipment supports, storage facilities, silos, etc., the technical solutions of the present invention are equally applicable to any structure with multiple load-bearing nodes, such as bridge bearing monitoring, building structural health diagnosis, wind power equipment condition assessment, and antenna support structure monitoring. The above embodiments are merely preferred embodiments of the present invention and are not intended to limit the invention. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A structural state group identification device, comprising: Multiple force sensing components are arranged at each load-bearing node of the structure to be monitored to sense the load and / or torque borne by each load-bearing node; The processing unit, which is communicatively connected to each of the force sensing components, is configured as follows: Based on the real-time force value distribution of each force sensing component and the reference force value distribution of the structure, the change in the force value distribution consistency state relative to the reference state is determined. Based on the changes and the characteristics of each force value relative to the reference force value, the state change type of the structure is identified; wherein, the overall state change refers to the state in which each force value changes synchronously relative to the reference force value and the changes do not exceed the preset change threshold, and the local anomaly refers to the state in which there are abnormal measurement points where the force value deviation exceeds the preset deviation threshold and the changes exceed the preset change threshold.

2. The apparatus according to claim 1, characterized in that, The amount of change is determined in the following way: The real-time force distribution consistency index is calculated based on the real-time force value data of each load-bearing node, and the benchmark force distribution consistency index is calculated based on the benchmark force value data of each load-bearing node; the ratio or difference between the real-time force distribution consistency index and the benchmark force distribution consistency index is taken as the change amount. The force distribution consistency index is at least one of the following: force variance, force standard deviation, force coefficient of variation, or force interquartile range.

3. The apparatus according to claim 1, characterized in that, The synchronous change refers to the consistent trend of the change of each force value and the satisfaction of a preset ratio condition. The preset ratio condition includes: the difference in the change amplitude of each force value is within a preset range, or each force value changes according to a fixed ratio.

4. The apparatus according to claim 1, characterized in that, The processing unit is further configured to: when there is an abnormal measuring point where the force value deviation exceeds the preset deviation threshold, but the change amount does not exceed the preset change threshold, mark the measuring point as an observation point and increase the data acquisition frequency; within the preset observation window, if the change amount exceeds the preset change threshold, upgrade to a local abnormality alarm, otherwise deactivate the observation state.

5. The apparatus according to claim 1, characterized in that, The processing unit is further configured to: when a local anomaly is determined, analyze the coupling relationship of force value change between the abnormal measuring point and its associated measuring point based on the spatial location of each load-bearing node and the pre-constructed load transfer relationship model; when the force value of the abnormal measuring point decreases and the force value of the associated measuring point increases, and the magnitude of the force value change satisfies the transfer ratio in the load transfer relationship model, determine the abnormal measuring point as the root source node.

6. The apparatus according to claim 1, characterized in that, The processing unit is further configured to: extract the historical trend of force value at each measuring point based on the force value time series of each force sensing component within a preset historical period, estimate the expected time when the force value deviation reaches the preset warning threshold, and output the estimated information including the expected time and maintenance suggestions.

7. The apparatus according to claim 1, characterized in that, The processing unit is further configured to: When environmental parameters are acquired, and the change does not exceed the preset change threshold, and the correlation between the force change and the change in environmental parameters exceeds the preset correlation threshold, the warning output is suppressed. When a change in overall state is determined, it is distinguished whether it is caused by long-term normal aging of the structure or by occasional environmental factors. If it is caused by long-term normal aging of the structure, the reference force value distribution is adjusted to the direction of the real-time force value distribution at a preset update rate. If it is caused by occasional environmental factors, the reference force value distribution is kept unchanged.

8. The apparatus according to claim 1, characterized in that, The processing unit is further configured to: extract the direction, magnitude and rate of change of force values ​​at each measuring point relative to the reference force value distribution of the real-time force value distribution, form a force value change feature vector, match it with a preset anomaly pattern library, and determine the cause type of the anomaly based on the matching result; the preset anomaly pattern library stores multiple force value distribution change feature patterns corresponding to different cause types.

9. The apparatus according to claim 1, characterized in that, The force sensing component is any one of strain sensing, piezoelectric sensing, capacitive sensing or optical sensing, and its arrangement is as follows: arranged between the load-bearing node and the support surface to sense axial load, mounted on the side of the structure to sense shear force or overturning moment, or suspended at the suspension point to sense axial tension.

10. A method for identifying structural state groups, comprising: Under the baseline healthy condition, obtain the baseline force value distribution at each load-bearing node and calculate the baseline force value distribution consistency index; During operation, the real-time force distribution at each load-bearing node is acquired, and the consistency index of the real-time force distribution is calculated. Based on the real-time force distribution consistency index and the benchmark force distribution consistency index, the change in the force distribution consistency state is determined. Based on the changes in the magnitude and the characteristics of the changes in each force value relative to the reference force value, the type of state change of the structure is identified.

Images