A method and system for measuring the perpendicularity of a shaped steel column

Abstract

This invention discloses a method and system for measuring the verticality of steel columns, belonging to the field of measurement method technology. This method constructs a sensor network comprising multiple types of sensors, including inclinometers, vibration sensors, and strain gauges, and designs a collaborative analysis module for data fusion analysis. The inclinometer provides direct evidence of verticality changes, the vibration sensor captures the dynamic characteristics of environmental and load excitations, and the strain gauge reflects changes in the stress state within the structure. When the inclinometer data initially indicates an anomaly, the system does not immediately issue an alarm but automatically calls up vibration and strain data within the same time window for joint analysis. Abnormal vibration modes are identified through vibration spectrum energy analysis, and stress abrupt changes are captured through strain trend analysis. By utilizing cross-validation of multiple data sources, the method overcomes the problem of false alarms easily caused by single-data diagnosis.

Description

Technical Field

[0001] This invention relates to the field of measurement method technology, specifically to a method and system for measuring the verticality of steel columns. Background Technology

[0002] In the field of building construction, steel columns are the main load-bearing components, and their verticality is a key indicator for measuring the construction quality and operational safety of the structure. Excessive verticality deviation will lead to the redistribution of internal forces in the structure, which may cause serious consequences such as local buckling, failure of connection nodes, and even overall instability. Therefore, it is necessary to measure the verticality of steel columns regularly or continuously. At present, the existing technologies mainly include two categories: one is the traditional manual measurement method, such as using theodolites, total stations, or plumb bobs for periodic fixed-point observation. This method relies on manual operation and is difficult to achieve real-time and continuous monitoring. The other is the automated monitoring method, which usually uses a single tilt sensor installed on the steel column to transmit the tilt angle data back to the monitoring center via wired or wireless means. When the angle exceeds a preset threshold, an alarm is triggered. However, existing technologies have the following problems in practical applications. First, monitoring methods that rely on tilt data are easily affected by environmental interference, such as temperature changes, instantaneous wind vibration, or vibration of construction machinery, which may produce false alarms. This is because a simple change in tilt angle may originate from structural damage or may only be caused by temporary loads or disturbances. The system lacks the ability to identify the cause of the anomaly. Moreover, the response mechanism of existing methods is relatively simple, usually adopting a fixed mode of alarming when the threshold is exceeded, without differentiating the processing according to the nature and severity of the anomaly. This may lead to overreaction to slight fluctuations, consuming operation and maintenance resources, or insufficient response to slowly developing dangerous situations. The response speed, efficiency, and accuracy of traditional methods need to be improved. In order to address the shortcomings of existing technologies, this invention provides a method and system for measuring the verticality of steel columns to solve the above problems. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a method and system for measuring the verticality of steel columns. By constructing a sensor network comprising various sensors such as inclinometers, vibration sensors, and strain gauges, and designing a collaborative analysis module for data fusion analysis, the inclinometer provides direct evidence of verticality changes. The vibration sensor captures the dynamic characteristics of environmental and load excitations, and the strain gauge reflects changes in the stress state within the structure. When the inclinometer data initially indicates an anomaly, the system does not immediately trigger an alarm but automatically calls upon vibration and strain data from the same time window for joint analysis. Abnormal vibration modes are identified through vibration spectrum energy analysis, and stress abrupt changes are captured through strain trend analysis. This multi-data cross-validation overcomes the problem of false alarms that are easily detected by single-data diagnosis. This allows the system to more accurately distinguish between temporary environmental disturbances, short-term construction impacts, and structural damage or instability precursors, thereby improving the accuracy and reliability of anomaly detection and reducing false alarm and missed alarm rates.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for measuring the verticality of steel columns, the method comprising the following steps: Step S1, acquire multiple types of sensor data: acquire monitoring data periodically uploaded by the sensor network deployed on the target steel column; the monitoring data includes at least a first type of data reflecting changes in verticality and a second type of data reflecting changes in structural state; Step S2, based on multi-type data collaborative judgment of anomalies: perform fusion analysis on the first type of data and the second type of data. When it is determined that there is a verticality anomaly based on the first type of data, combine the second type of data in the corresponding time window to confirm the verticality anomaly and generate an anomaly judgment result. The anomaly judgment result includes at least an anomaly type identifier and a quantitative parameter used to characterize the severity of the anomaly. Step S3: Execute the hierarchical response corresponding to the anomaly determination result: Based on the numerical range of the quantization parameter, trigger different levels of response actions from data recording to security linkage.

[0005] Preferably, the first type of data is tilt angle or angle change data collected by the tilt sensor; the second type of data includes vibration data collected by the vibration sensor and / or strain data collected by the strain gauge.

[0006] Preferably, step S2 includes: Step S21, Initial judgment of verticality anomaly: The first type of data is judged according to the preset tilt angle threshold or angle change rate threshold. If the threshold is exceeded, an anomaly trigger signal is generated. Step S22, Multi-source data collaborative verification: In response to the abnormal triggering signal, extract the second type of data within the same time window as the first type of data, perform spectral energy analysis on the vibration data to obtain the main vibration frequency and amplitude characteristics, and / or perform trend analysis on the strain data to obtain the strain abrupt change gradient; Step S23, type and degree determination: Match the dominant frequency, amplitude characteristics and / or strain abrupt change gradient with a preset rule base of different anomaly types and quantization parameters, and output the anomaly determination result.

[0007] Preferably, the quantification parameter is a comprehensive risk score; in step S3, the graded response includes: If the comprehensive risk score is within the first numerical range, then local data recording and trend analysis are performed; If the comprehensive risk score is in the second value range, an early warning notification is sent to the remote management terminal, and the data acquisition frequency of the relevant sensors is increased. If the comprehensive risk score is in the third value range, a linkage control command is sent to the regional security system. The linkage control command is used to trigger an audible and visual alarm or control related devices to enter a safe state.

[0008] Preferably, the sensor network is a wireless network and / or a wired network.

[0009] Preferably, the monitoring data also includes a third type of data, which is settlement monitoring data deployed at the location of the target steel column foundation; in step S2, when the verticality abnormality is confirmed and an abnormality judgment result is generated, the third type of data is further fed back to facilitate personnel analysis and judgment.

[0010] Preferably, the method further includes step S4: constructing and updating a digital twin model of the target steel column based on historical monitoring data; in step S2, comparing the real-time acquired monitoring data with the predicted state of the digital twin model to assist in anomaly determination.

[0011] Preferably, the method further includes step S5: optimizing the preset threshold and the matching rules in the abnormal pattern library based on multiple anomaly determination results and subsequent manual handling feedback information.

[0012] Preferably, when the response action is to increase the data acquisition frequency, the method further includes: after continuously monitoring that the comprehensive risk score has fallen below the first numerical range for a preset time, automatically restoring the data acquisition frequency to the initial frequency and generating a closed-loop processing log for the abnormal event.

[0013] This invention also discloses a steel column verticality measurement system for implementing the steel column verticality measurement method, the system comprising: A sensor network, deployed on the target steel column and / or the foundation of the steel column, is used to collect first-type data and second-type data; An edge gateway, which is communicatively connected to the sensor network, is used to aggregate and upload the monitoring data; The data receiving module is used to acquire data from multiple types of sensors; The collaborative analysis module is used to collaboratively determine anomalies based on multiple types of data, and the collaborative analysis module integrates a data fusion analysis engine. The early warning response module is used to execute tiered responses corresponding to the anomaly determination results; The client terminal is used to receive and display early warning information and system reports from the cloud server.

[0014] The technical effects and advantages of this invention are as follows: 1. This method for measuring the verticality of steel columns constructs a sensor network comprising various sensors, including inclinometers, vibration sensors, and strain gauges. A collaborative analysis module is designed for data fusion analysis. The inclinometer provides direct evidence of verticality changes, the vibration sensor captures the dynamic characteristics of environmental and load excitations, and the strain gauge reflects changes in the stress state within the structure. When the inclinometer data initially indicates an anomaly, the system does not immediately trigger an alarm. Instead, it automatically calls upon vibration and strain data from the same time window for joint analysis. Abnormal vibration modes are identified through vibration spectrum energy analysis, and stress abrupt changes are captured through strain trend analysis. This multi-data cross-validation overcomes the problem of false alarms caused by single-data diagnosis. This allows the system to more accurately distinguish between temporary environmental disturbances, short-term construction impacts, and structural damage or instability precursors, thereby improving the accuracy and reliability of anomaly detection and reducing false alarm and missed alarm rates.

[0015] 2. This method for measuring the verticality of steel columns, by setting a rule base for anomaly types and quantitative parameters, and introducing a comprehensive risk score as a quantitative parameter, achieves accurate classification and severity grading of abnormal events. The collaborative analysis module matches the feature parameters obtained from the fusion analysis with the preset rule base, not only outputting abnormal signals, but also further determining the type of anomaly and calculating a comprehensive risk score, transforming ambiguous abnormal states into judgment results with clear types and quantitative levels. The early warning response module can execute graded response actions according to different numerical ranges of the comprehensive risk score, achieving optimal resource allocation: low-risk fluctuations are only recorded and observed, medium-risk events are monitored more frequently and warnings are issued, and high-risk emergency situations are immediately triggered to activate equipment linkage and alarms, thereby improving the level of safety management.

[0016] 3. Based on the method of measuring the verticality of steel columns, a long-term monitoring system was constructed. The flexible deployment of the sensor network enables it to adapt to different scenarios such as new projects or renovations of existing buildings. The digital twin model provides a reference benchmark and early warning capability for anomaly detection by comparing the predicted state of the physical entity with that of the virtual model, ensuring the reliability and effectiveness of long-term monitoring. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of the overall method of the present invention; Figure 2 This is a system architecture diagram of the present invention; Figure 3 This is the anomaly detection logic diagram for the present invention; Figure 4 This is the hierarchical response logic diagram of the present invention. Detailed Implementation

[0019] 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, and 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.

[0020] This embodiment discloses a method for measuring the verticality of steel columns, according to the appendix. Figure 1 To be continued Figure 4 As shown, the present invention includes a method and a corresponding system, the flowchart of which is attached. Figure 1 As shown, it mainly includes: Step S1: Acquire various types of monitoring data uploaded by the sensor network; Step S2: Perform fusion analysis on multiple types of data, identify anomalies, and generate a judgment result with quantified parameters; Step S3: Trigger a graded response action based on the quantization parameters; The system architecture of this invention is shown in the appendix. Figure 2 As shown, it includes a sensor network and edge gateway deployed on-site, as well as a collaborative analysis module, an early warning response module, and a client terminal located on a server or in the cloud.

[0021] Furthermore, in step S1, the construction of the sensor network is the foundation for implementation. The sensor network can be flexibly networked using wireless and / or wired networks to adapt to different construction sites or existing building deployment environments. The network includes multiple types of sensor nodes: at least a tilt sensor for collecting the first type of data, preferably a high-precision dual-axis or single-axis MEMS tilt sensor, which is directly installed on the monitoring section of the steel column, usually located at the top, middle, or critical position considered to be prone to deformation, for continuously or periodically measuring the tilt angle or angle change of the column relative to the gravity reference; and sensors for collecting the second type of data, such as vibration sensors and strain gauges. The strain gauges are preferably resistance strain gauges or fiber optic strain sensors. The vibration sensor is used to capture vibration signals caused by wind load, equipment operation, construction activities, or abnormal deformation of the structure, while the strain gauge is used to measure the strain state of the key section of the steel column, preferably such as the column base or near the connection node, reflecting its stress changes. All sensors are aggregated, preliminarily processed, and uploaded to the back-end platform through an edge gateway. The preliminary processing includes analog-to-digital conversion and data packaging.

[0022] Furthermore, the detailed logic of the fusion analysis in step S2 is attached. Figure 3 As shown, step S21 is executed first to perform a preliminary judgment on the first type of data. The system presets an absolute threshold for tilt angle (e.g., the cumulative tilt angle exceeds H / 250, where H is the column height) and a threshold for the rate of angle change (e.g., the angle increment exceeds the limit per unit time). When the tilt data calculated in real time or periodically exceeds either threshold, the system generates a preliminary abnormality trigger signal. This step serves as the first filter to initially identify potential verticality problems.

[0023] Specifically, after generating an abnormal trigger signal, the system does not immediately alarm, but enters the collaborative confirmation stage in step S22. The system extracts the second type of data within the same time window associated with the trigger time. For vibration data, spectral energy analysis is performed: the time-domain vibration signal is converted into a frequency-domain signal through Fast Fourier Transform (FFT), and the dominant frequency and corresponding amplitude characteristics are extracted. The dominant frequency is the frequency component with the most concentrated energy. For example, low-frequency, high-amplitude vibration may originate from foundation settlement or overall instability; high-frequency, specific-frequency vibration may be related to resonance from nearby mechanical operations. For strain data, trend analysis is performed: the slope or higher-order derivative of the strain data is calculated, and the strain abrupt change gradient is obtained to determine whether there is stress concentration or local buckling. For example, if the column base strain continues to increase rapidly without new load, it may indicate foundation loosening or column bending.

[0024] Specifically disclosed, step S23 involves pattern matching and judgment output. The system maintains a preset rule base for anomaly types and quantification parameters. This rule base defines the mapping relationship between the feature vectors of different anomaly patterns and the "comprehensive risk score," for example: Rule 1: If the tilt exceeds the limit and the dominant vibration frequency is low frequency 1-3Hz and the amplitude is continuously large and the strain at the column base shows a stable increasing trend, then it is matched to the foundation settlement or slip anomaly type, and the comprehensive risk score = f1 (tilt angle) + f2 (amplitude) + f3 (strain gradient); Rule 2: If the tilt rate exceeds the limit and the vibration spectrum exhibits wide-bandwidth, impact-like characteristics, and a sudden change in strain occurs at a certain node, it may be matched as an anomaly type of instantaneous impact or collision damage. The quantitative parameter comprehensive risk score is a normalized value that integrates the deviation degree and weight of various characteristic parameters to quantify the severity and urgency of the abnormal event.

[0025] It is particularly important to emphasize that, in order to further improve the comprehensiveness of the diagnosis, the monitoring data can also include a third type of data, namely, the settlement monitoring point data deployed at the target steel column foundation location, such as hydrostatic level data. In the anomaly judgment process of step S2, after the system initially determines that the verticality is abnormal, the foundation settlement data at the corresponding time can be fed back to the client terminal for managers to conduct a more comprehensive analysis and judgment. For example, if the tilting anomaly is accompanied by significant uneven foundation settlement, the root cause can be more clearly pointed to the foundation problem.

[0026] It is particularly important to emphasize that the method also includes advanced functional step S4: Based on historical monitoring data, a digital twin model of the target steel column is constructed and continuously updated. This model is established through finite element analysis (FEA) or machine learning algorithms and can simulate the structural response under specific loads and environmental conditions. In the real-time monitoring in step S2, the actual monitoring data can be compared with the predicted state of the digital twin model under the same input conditions in real time. If the actual data deviates significantly from the predicted value, even if the primary threshold is not triggered immediately, it can be used as an auxiliary criterion to warn of potential risks in advance, or to verify the rationality of the current anomaly judgment result.

[0027] Furthermore, the hierarchical response mechanism in step S3 is as follows: Figure 4 As shown, the system automatically triggers response actions at different levels based on the numerical range of the comprehensive risk score output in step S23. For the first numerical range, for example, a score of 0-40, it is determined to be a slight abnormality or normal fluctuation. The system only performs local data recording and strengthens data trend analysis, without issuing external alarms. The second numerical range, for example, a score of 41-70: is judged as a moderate risk. The system will automatically send a yellow warning notification to the remote management terminal, and report the anomaly type, risk score and key data in detail. At the same time, it will instruct the edge gateway to increase the data acquisition frequency of the relevant sensors in order to obtain data with higher time resolution for close monitoring. The third numerical range, for example, a score of 71-100, indicates a high-risk or emergency situation. The system immediately sends a linkage control command to the regional safety system, such as the construction site's audible and visual alarm system, equipment control system, and emergency broadcast system. The command can trigger a high-intensity audible and visual alarm to warn on-site personnel to evacuate; or control related equipment to enter a safe state, such as suspending hoisting operations in adjacent areas or shutting down related power equipment to prevent the accident from escalating.

[0028] Furthermore, the method also includes a self-learning optimization step S5. The system records the results of each anomaly judgment, the trigger response, and the feedback information of subsequent manual on-site verification and handling, such as confirming false alarms, confirming the anomaly type and cause, and recording the handling measures. Using this feedback data, the system periodically or based on a certain number of new feedback samples, through machine learning algorithms such as decision tree optimization and Bayesian network updates, iteratively optimizes the preset threshold in step S2, the feature matching rules in the rule base, and the weight coefficients of the comprehensive risk score. This enables the system to continuously adapt to specific engineering environments, reduce false alarms and false negatives, and gradually improve diagnostic accuracy over time.

[0029] It is particularly important to emphasize that, to ensure the rational use of system resources, when the data collection frequency is increased due to the risk score entering the second range, the system has the ability to adaptively recover. For example, after continuously monitoring that the risk score has fallen back to below the first value range and remained stable for a preset period of time, the system will automatically instruct the sensor network to restore the collection frequency to the initial normal frequency and generate a closed-loop processing log that includes the entire process of abnormal triggering, upgraded monitoring, risk mitigation, and frequency recovery, providing post-event auditing and analysis.

[0030] Example 1: This example uses the monitoring of the core tube steel column of a high-rise building during the construction phase as an example, combined with the attached... Figure 1 To be continued Figure 4 The workflow is explained in detail below: Dual-axis inclinometers are installed at the top and middle of the key steel columns (corner columns and columns in complex stress areas) of the core tube, strain gauges are installed at the column bases, vibration sensors are installed on adjacent floor slabs, and a sensor network is built through a wireless LoRa network, with data collected by edge gateways deployed in the floors.

[0031] The system periodically (e.g., every minute) acquires tilt data. If, during the hoisting of a heavy curtain wall unit on one side, the system detects that the rate of change of the tilt angle at the top of column A-12 exceeds the preset threshold for the rate of change of construction activity impact within 10 minutes, an abnormal trigger signal is generated.

[0032] The system immediately retrieved the vibration and strain data of column A-12 within the time window. Vibration spectrum analysis showed that the main vibration frequency matched the operating frequency of the hoisting equipment, and the amplitude was large but it was an instantaneous impact. The strain at the column base fluctuated slightly but did not form a continuous growth gradient. The rule base matched this pattern as an instantaneous deviation caused by construction dynamic load. The comprehensive risk score was calculated to be 55, and it entered the second interval.

[0033] The system sent a yellow alert to the project manager's mobile app: the instantaneous tilt rate of column A-12 exceeded the limit, suspected to be affected by hoisting operations, with a risk score of 55. At the same time, the system instructed to increase the sensor acquisition frequency of column A-12 and the two adjacent columns from 1 time / minute to 1 time / 10 seconds.

[0034] After the hoisting operation was completed, continuous monitoring showed that the tilt angle tended to stabilize, the vibration returned to the background value, and the risk score dropped to 35 after 20 minutes. The system automatically restored the acquisition frequency and generated a log: "At xx time, column A-12 triggered a momentary tilt warning due to hoisting operations. Monitoring has shown that it has stabilized and the frequency has been restored." Example 2: This example uses the monitoring of steel columns near heavy equipment during the operation of an industrial plant as an example, combined with the attached... Figure 1 To be continued Figure 4 The workflow is explained in detail below: A wired sensor network of RS485, including inclinometers, vibration sensors, and strain gauges at multiple locations on the column, is deployed on the load-bearing steel columns near the large reciprocating compressor in the factory building.

[0035] The system analyzes the tilt trend daily. If the system finds that the tilt angle of column B-05 shows a slow but continuous unidirectional growth trend over several weeks, even though the daily change does not exceed the rate threshold, the cumulative value is close to the absolute angle threshold.

[0036] The system proactively conducted in-depth analysis of recent data. Vibration data analysis revealed that, in addition to the fixed frequency of equipment operation, there were continuous low-frequency weak vibration components. Strain data analysis showed that the compressive strain on one side of the column base was increasing slowly and continuously. When the digital twin model was called and the current equipment vibration load was input, the model predicted the column response to be much smaller than the actual monitored tilt and strain values, indicating a significant deviation.

[0037] Combining the third type of data, the foundation settlement monitoring point of the column was examined, and it was found that there was slight but uneven settlement during the same period. Based on the rule base's comprehensive analysis of slow tilting, low-frequency abnormal vibration, continuous strain growth, model deviation, and foundation settlement characteristics, it was determined that uneven foundation settlement caused the column to gradually tilt and redistribute internal forces, and the comprehensive risk score was calculated to be 68.

[0038] The system immediately sent an orange alert report to the equipment maintenance department and the structural safety department, clearly pointing out the suspected basic problem and recommending professional testing. At the same time, all monitoring points in the area were included in the high-frequency monitoring mode.

[0039] Subsequent detection by artificial ground-penetrating radar confirmed localized soil softening beneath the foundation, thus providing a successful early warning for this anomaly. The data pattern was recorded by the system and used to optimize the recognition sensitivity and scoring rules in the rule base.

[0040] Example 3: This example uses the monitoring of steel columns near the foundation pit during the renovation of an existing building as an example. The workflow is as follows: A monitoring system was deployed on the steel columns of the adjacent building's ground floor on the excavation side of the foundation pit. If the system detected a sharp increase in the tilt angle of column C-03 within a short period of time, triggering an anomaly, collaborative analysis revealed that the vibration signal was weak, ruling out impact from large machinery. However, the column strain exhibited drastic abrupt changes in gradient across multiple cross sections, and the foundation settlement data simultaneously showed significant subsidence. The rule base quickly matched this to a rapid slippage or instability mode caused by soil stress release, with the risk score reaching the third interval of 92. The system immediately triggered the highest level of response: the audible and visual alarms at the construction site sounded continuously, and an emergency shutdown command was sent to the foundation pit monitoring platform to suspend all excavation operations. At the same time, an alarm was alerted to the emergency command center. This rapid response bought time for personnel evacuation and emergency support, preventing the accident from escalating.

[0041] Example 4: This example uses monitoring in response to sudden external events as an example. The workflow is as follows: After the earthquake, all building monitoring points in the area where this system is deployed simultaneously enter emergency analysis mode. The system not only checks the verticality of individual columns, but also quickly performs cluster analysis on the tilt patterns, vibration spectrum consistency, and strain distribution of multiple columns within the same structure. For a building in Zone D, if the system finds that multiple columns tilt in tandem and the vibration spectrum contains significant seismic characteristic frequencies, but the strain does not show localized destructive concentration, the rule base determines it as overall swaying, the structure is in an elastic response state, and the risk score is 50. The system sends a notification to the administrator that the structure has withstood the earthquake and the overall response is normal, suggesting a follow-up inspection. For another building, if the system finds that a column tilts abnormally and is accompanied by localized high-frequency vibrations and strain spikes, it determines that there may be local component damage and marks it as a high-priority inspection target.

[0042] Example 5: This example uses the long-term self-learning and optimization process of the system as an example. The workflow is as follows: In the early stages of system operation, tilting fluctuations caused by certain specific wind vibrations might be falsely reported as moderate risk due to an incomplete rule base. Each time a false alarm or wind vibration was manually confirmed on the client side, the feedback information was recorded. After the system accumulated many similar cases, the self-learning module analyzed and found that the vibration spectrum in such false alarm events had specific wind-induced vibration broadband characteristics and was strongly correlated with anemometer data. Therefore, the module automatically suggested and optimized the rules: after tilting is triggered, if the vibration spectrum matches the wind vibration characteristics and the wind speed is high, the risk score weight of the event is automatically reduced, or it is classified as an environmental interference subclass, and only recorded without escalating the alarm. After several quarters of operation, the system's ability to distinguish between common environmental interferences and real structural anomalies has improved, and the false alarm rate has decreased.

[0043] In summary, this method for measuring the verticality of steel columns constructs a sensor network comprising multiple types of sensors, including inclinometers, vibration sensors, and strain gauges. A collaborative analysis module is designed for data fusion analysis. The inclinometer provides direct evidence of verticality changes, the vibration sensor captures the dynamic characteristics of environmental and load excitations, and the strain gauge reflects changes in the stress state within the structure. When the inclinometer data initially indicates an anomaly, the system does not immediately trigger an alarm but automatically calls upon vibration and strain data from the same time window for joint analysis. Abnormal vibration modes are identified through vibration spectrum energy analysis, and stress abrupt changes are captured through strain trend analysis. By utilizing cross-validation of multiple data sources, the method overcomes the problem of false alarms that are easily detected by single-data diagnosis.

[0044] Moreover, by setting a rule base for anomaly types and quantification parameters, and introducing a comprehensive risk score as a quantification parameter, this invention achieves accurate classification and severity grading of abnormal events. The collaborative analysis module matches the feature parameters obtained from the fusion analysis with the preset rule base, not only outputting anomaly signals, but also further determining the anomaly type identifier and calculating a comprehensive risk score, transforming ambiguous anomaly states into judgment results with clear types and quantification levels. The early warning response module can execute graded response actions according to the different numerical ranges of the comprehensive risk score, achieving optimal resource allocation.

[0045] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for measuring the verticality of steel columns, characterized in that, The method includes the following steps: Step S1: Obtain monitoring data periodically uploaded by the sensor network deployed on the target steel column; the monitoring data includes at least a first type of data reflecting changes in verticality and a second type of data reflecting changes in structural state; Step S2: Perform fusion analysis on the first type of data and the second type of data. When it is determined that there is a verticality anomaly based on the first type of data, combine the second type of data in the corresponding time window to confirm the verticality anomaly and generate an anomaly determination result. The anomaly determination result includes at least an anomaly type identifier and a quantitative parameter used to characterize the severity of the anomaly. Step S3: Based on the numerical range of the quantization parameter, trigger different levels of response actions, from data recording to security linkage.

2. The method for measuring the verticality of steel columns according to claim 1, characterized in that, The first type of data is the tilt angle or angle change data collected by the tilt meter sensor; the second type of data includes vibration data collected by the vibration sensor and / or strain data collected by the strain gauge.

3. The method for measuring the verticality of steel columns according to claim 2, characterized in that, Step S2 includes: Step S21: Based on the preset tilt angle threshold or angle change rate threshold, the first type of data is judged. If the threshold is exceeded, an abnormal trigger signal is generated. Step S22: In response to the abnormal trigger signal, extract the second type of data within the same time window as the first type of data, perform spectral energy analysis on the vibration data to obtain the main vibration frequency and amplitude characteristics, and / or perform trend analysis on the strain data to obtain the strain abrupt change gradient; Step S23: Match the dominant frequency, amplitude characteristics and / or strain abrupt change gradient with a preset rule base of different anomaly types and quantization parameters, and output the anomaly determination result.

4. The method for measuring the verticality of steel columns according to claim 3, characterized in that, The quantitative parameter is a comprehensive risk score; In step S3, the graded response includes: If the comprehensive risk score is within the first numerical range, then local data recording and trend analysis are performed; If the comprehensive risk score is in the second value range, an early warning notification is sent to the remote management terminal, and the data acquisition frequency of the relevant sensors is increased. If the comprehensive risk score is in the third value range, a linkage control command is sent to the regional security system. The linkage control command is used to trigger an audible and visual alarm or control related devices to enter a safe state.

5. The method for measuring the verticality of steel columns according to claim 1, characterized in that, The sensor network is a wireless network and / or a wired network.

6. The method for measuring the verticality of steel columns according to claim 2, characterized in that, The monitoring data also includes a third type of data, which is settlement monitoring data deployed at the location of the target steel column foundation; in step S2, when the verticality abnormality is confirmed and an abnormality judgment result is generated, the third type of data is further fed back to facilitate personnel analysis and judgment.

7. The method for measuring the verticality of steel columns according to claim 2, characterized in that, The method further includes step S4: constructing and updating a digital twin model of the target steel column based on historical monitoring data; in step S2, the real-time acquired monitoring data is compared with the predicted state of the digital twin model to assist in anomaly determination.

8. The method for measuring the verticality of steel columns according to claim 3, characterized in that, The method further includes step S5: optimizing the preset threshold and the matching rules in the abnormal pattern library based on multiple anomaly determination results and subsequent manual handling feedback information.

9. The method for measuring the verticality of steel columns according to claim 4, characterized in that, When the response action is to increase the data collection frequency, the method further includes: after continuously monitoring that the comprehensive risk score has fallen below the first numerical range for a preset time, automatically restoring the data collection frequency to the initial frequency and generating a closed-loop processing log for the abnormal event.

10. A system for measuring the verticality of steel columns, characterized in that, For implementing the steel column verticality measurement method according to any one of claims 1-9, the system comprises: A sensor network, deployed on the target steel column and / or the foundation of the steel column, is used to collect first-type data and second-type data; An edge gateway, which is communicatively connected to the sensor network, is used to aggregate and upload the monitoring data; The data receiving module is used to acquire data from multiple types of sensors; The collaborative analysis module is used to collaboratively determine anomalies based on multiple types of data, and the collaborative analysis module integrates a data fusion analysis engine. The early warning response module is used to execute tiered responses corresponding to the anomaly determination results; The client terminal is used to receive and display warning information from the warning response module.

Images