A safety monitoring system for steel truss columns of overhead climbing frame with wall support integrated with big data analysis

Abstract

This invention discloses a safety monitoring system for steel truss columns supporting climbing scaffolds in elevated floors, integrating big data analysis. It relates to the field of construction monitoring technology and aims to solve the problem of misjudgment caused by construction disturbances. The system extracts behavioral labels from construction behavior control signals, constructs a semantic mapping relationship between construction behavior and structural response, generates a behavior-driven dataset, extracts disturbance change trends through node response tensor and time difference analysis, integrates behavioral labels to construct a structural state vector, establishes a structural behavior distribution model, analyzes the deviation between the current state and the model response, and identifies the spatial propagation path of anomalies using a structural layout diagram. It employs graph clustering to extract high-risk anomaly clusters, analyzes structural state changes and construction behavior in anomaly areas, triggers different monitoring response strategies, and achieves structural risk identification and control during construction, improving the efficiency of structural safety monitoring during the floor construction phase.

Description

Technical Field

[0001] This invention relates to the field of construction monitoring technology, and more specifically, to a safety monitoring system for steel truss columns of elevated climbing scaffolding with integrated big data analysis. Background Technology

[0002] In the construction of high-rise buildings, the elevated climbing scaffold serves as the main formwork support and safety protection device. It is connected to the steel truss columns through wall-mounted supports and rises and falls periodically with the construction progress. The rising and falling behavior of the climbing scaffold not only changes the boundary conditions between itself and the main structure, but also causes abrupt changes in the local stress state of the steel truss columns. With the development of big data analysis and intelligent sensing technology, the method of integrating structural physical state, construction behavior characteristics and time series analysis has become an important direction for intelligent monitoring of high-rise buildings.

[0003] The shortcomings of existing technologies: For structural safety monitoring systems of climbing scaffold wall supports and steel truss columns in elevated floors, the judgment method is generally based on stress or displacement thresholds. It lacks the ability to dynamically identify and integrate the construction behavior status. In actual construction, operations such as climbing scaffold raising and lowering can cause rapid changes in the structural stress boundary conditions, resulting in local stress mutations or displacement responses. However, the structural responses caused by such construction behaviors are often misjudged as abnormal by the system, or conversely, the real anomalies are masked because they are not separated from the construction behavior. Since there is no correlation mapping between structural data and construction behavior, existing systems cannot effectively distinguish between expected mutations and unexpected anomalies in the construction process. They also lack the logical scheduling ability based on multi-source data modeling and state-driven logic, resulting in insufficient accuracy of monitoring results, delayed risk identification, and a single response strategy, which makes it difficult to meet the safety monitoring needs in complex dynamic construction environments. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the following solution is proposed to solve the problem of misjudgment caused by construction disturbance in the above-mentioned background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A safety monitoring system for steel truss columns supporting climbing scaffolding in elevated floors, which integrates big data analysis, includes an elevated floor monitoring and data acquisition module, a structural state change module, a structural area anomaly identification module, and a response strategy module. The modules are connected by signals.

[0007] The elevated floor monitoring and data acquisition module is used to collect multimodal response information of steel truss columns and wall-attached support areas. It extracts behavior labels from the collected data according to the action type and state switching point, establishes a mapping relationship between construction behavior and structural response, and generates behavior-driven semantic structural response data.

[0008] The structural state change module is used to construct node response tensors based on structural response data, extract the perturbation change trend of structural nodes by combining time difference, and inject behavioral labels to generate node activation functions in the process of structural node state expression, thereby constructing the structural state vector of structural nodes.

[0009] The structural anomaly identification module is used to establish a structural behavior distribution model based on construction behavior conditions, identify the degree of anomaly deviation based on the response difference between the current state and the model, analyze the spatial propagation characteristics of anomalies based on the structural layout diagram, and extract anomaly cluster regions using graph clustering.

[0010] The response strategy module is used to acquire the structural state evolution characteristics within high-risk anomalous clumps in the anomalous clump region, analyze the temporal coordination information between structural response behavior and construction process, and trigger different monitoring response strategies based on the analysis results.

[0011] In a preferred embodiment, multimodal response information of the steel truss columns and wall-mounted support areas is collected, and behavioral labels are extracted from the collected data based on action type and state transition points. The specific steps are as follows:

[0012] Multimodal response information of steel truss columns and wall-mounted supports is collected. The monitoring and collection units include the upper and lower sections of the steel truss columns and the node connections, the rigid contact surface and sliding area of ​​the wall-mounted supports, and the lifting and lowering operation process of the climbing frame.

[0013] Sensors are deployed in the data acquisition unit, including strain gauges placed in the upper, middle and lower sections of the steel truss column, to obtain the strain response of the column under vertical load and horizontal disturbance.

[0014] Displacement sensors are installed between the wall support and the structural wall to monitor micro-slippage, loosening, and relative displacement at the wall attachment points.

[0015] High-frequency accelerometers are installed at the node connections to capture rapid impact and vibration responses during the climbing frame's lifting process;

[0016] The behavior acquisition device is connected to the climbing scaffold electrical control system to collect lifting execution commands, action start and end signals, and lifting displacement change data.

[0017] Set a unified global sampling time series, and resample or interpolate all collected data based on the least common multiple sampling step size.

[0018] The collected data is analyzed to extract action types and state transition points, forming behavior state labels. These labels include whether the current state is in the lifting, steady state, unlocked, or stopped state.

[0019] In a preferred embodiment, a mapping relationship between construction behavior and structural response is established, and behavior-driven semantic structural response data is generated. Specific steps include:

[0020] Establish a mapping relationship between construction behavior and structural response units. The mapping relationship is determined based on the structural layout diagram, construction equipment layout and on-site experience. It is represented in the form of a response influence diagram in the system, which maps the behavior state to a set of structural response nodes.

[0021] The time-synchronized collected data, behavioral tags, and structural response node sets are merged to form structural response data;

[0022] The structural response data includes the modal response values ​​of all controlled nodes at the current time point, the current behavior label, the active response area corresponding to the behavior, the node number and physical location data.

[0023] In a preferred embodiment, a node response tensor is constructed based on structural response data, and the perturbation change trend of the structural nodes is extracted by combining time difference. Behavioral labels are injected into the structural node state representation process to generate a node activation function, thereby constructing the structural node's structural state vector. The specific steps are as follows:

[0024] The collected data are integrated into a modal response tensor.

[0025] The time difference of each response value in the modal response tensor is calculated, the velocity and acceleration that change with time are extracted, and the modal dimension of adjacent time points is given to determine the response mutation function.

[0026] Trend feature sequences are extracted from the raw collected data, and structural state evolution modeling is performed based on time.

[0027] A behavior label is introduced to drive the response of each node, and the structural state vector of the node is constructed.

[0028] In a preferred embodiment, a behavior label is introduced to drive the response of each node, and a node structure state vector is constructed. The specific steps are as follows:

[0029] Get every time Behavioral tags ;

[0030] Find the influence structure region corresponding to the behavior label Obtain the set of nodes whose behavior causes disturbance;

[0031] If node Then set the node activation coefficient to . Otherwise, set the node activation coefficient. ,in, This is the effect strength parameter of the current node relative to the behavioral state. The area affected by the response to the behavior; Where T represents the number of time steps in which the construction activity continues. , These are the times of node i at time i. and The total structural response;

[0032] By combining the modal response tensor, trend perturbation function, and node activation function, a node at time t is constructed. Structure state vector: ,in, For modal response tensors, This represents a trend disturbance characteristic.

[0033] In a preferred embodiment, a structural behavior distribution model is established based on construction behavior conditions, and the degree of abnormal deviation is identified based on the response difference between the current state and the model. The specific steps are as follows:

[0034] Establish a structural behavior distribution model under normal operating conditions, perform deviation judgment, and set the state distribution function as follows: ,in, Indicates the time of the i-th structural node. The state vector, Indicates time Corresponding behavioral tags For specific construction activities; Indicates that the construction activity is in a state of... The state vector of node i at state i The conditional probability distribution;

[0035] Real-time deviation detection and abnormal deviation identification of state vectors are performed. For each structural node, the typical response trajectory of the node under this behavior is extracted based on the behavior label at the current moment, and the deviation strength score between the current state vector and the reference vector is calculated. ,in, This represents the m-th modal dimension of the state vector; Indicates behavioral conditions The reference trajectory value in the m-th dimension; For deviation mapping function;

[0036] Actual state response value Through real-time sensors The reference trajectory value is obtained by collecting data at all times and then standardizing and mapping the state representation. In the construction activity In the scenario, the normal state value of the m-th modal dimension obtained through the historical structural state trajectory is used as the baseline for deviation judgment;

[0037] Deviation strength score is an indicator of the anomalous significance of the node in its current behavioral state;

[0038] Modeling is performed using a Markov latent variable graphical model to establish a state transition graph for each node and to establish a latent variable space for the response trend of each modality dimension. Under each behavior label condition, the probability of node state trajectory transition is inferred.

[0039] In a preferred embodiment, the spatial propagation characteristics of anomalies are analyzed based on the structural layout diagram, and graph clustering is used to extract anomalous cluster regions. The specific steps are as follows:

[0040] Perform spatial clustering analysis and propagation trend identification of structural anomalies, that is, perform spatial clustering analysis on the deviation values ​​of all nodes to identify local anomaly clustering areas and propagation paths;

[0041] The structural layout diagram is converted into a structural atlas, where nodes represent monitoring points and edges represent component connections.

[0042] Define the degree of abnormal deviation on the structural layout drawing: ,in , These represent the deviation strength scores for nodes i and j, respectively. This represents the weight of the connection edge between the i-th node and the j-th node in the structural layout diagram; , indicates that node i and node j have a direct connection or structural coupling relationship in the structure; E represents the set of edges in the structural layout diagram;

[0043] By combining graph clustering algorithms, abnormal clusters are extracted from the off-center areas, the affected areas are identified, and high-risk abnormal clusters are determined.

[0044] In a preferred embodiment, a graph clustering algorithm is used to extract anomalous clusters from the deviation region, identify the affected region, and determine high-risk anomalous clusters. The specific steps are as follows:

[0045] Real-time calculation of the degree of abnormal deviation of each node in the structural layout diagram;

[0046] The degree of abnormal deviation of nodes is integrated into the structural connection diagram to form a composite structural diagram that has both structural connection relationships and degree of abnormal behavior.

[0047] By combining the structural connection strength with the abnormal deviation differences between each node, an anomaly similarity map is constructed and similarity analysis is performed;

[0048] Graph clustering techniques are used to partition the entire anomaly similarity graph, identifying sets of nodes that are physically adjacent to each other and highly consistent in their anomaly behavior.

[0049] After clustering is completed, the degree of abnormal deviation of all nodes in each cluster region is calculated. If the degree of abnormal deviation of nodes in a region shows an upward trend, the region is marked as a high-risk abnormal cluster.

[0050] In a preferred embodiment, the method for acquiring the structural state evolution characteristics within high-risk anomalous clumps in an anomalous clumping region, analyzing the temporal coordination information between structural response behavior and the construction process, and triggering different monitoring and response strategies based on the analysis results, includes the following specific steps:

[0051] Obtain the structural state evolution characteristics within high-risk anomalous cluster regions and analyze the temporal coordination information between structural response behavior and construction process;

[0052] Temporal synergy information includes the structural intervention urgency index and the monitoring strategy evolution intensity index;

[0053] The structural intervention urgency index is used to represent the abruptness of structural disturbances, the hysteresis of response recovery, and the temporal correlation between disturbances and construction activities.

[0054] The monitoring strategy evolution intensity index is used to indicate the urgency of whether the current monitoring system strategy needs adaptive adjustment;

[0055] Set thresholds for the intensity of monitoring strategy evolution and the urgency of structural intervention;

[0056] The threshold for the intensity of monitoring strategy evolution and the threshold for the urgency of structural intervention were compared and analyzed with the intensity index of monitoring strategy evolution and the urgency index of structural intervention, respectively.

[0057] In a preferred embodiment, the monitoring strategy evolution intensity threshold and the structural intervention urgency threshold are compared and analyzed with the monitoring strategy evolution intensity index and the structural intervention urgency index, respectively. The specific steps are as follows:

[0058] When the structural intervention urgency index is lower than the structural intervention urgency threshold and the monitoring strategy evolution intensity index is lower than the monitoring strategy evolution intensity threshold, the structure is in an acceptable risk range, and the current observation strategy and response remain silent.

[0059] When the structural intervention urgency index is higher than the structural intervention urgency threshold and the monitoring strategy evolution intensity index is lower than the monitoring strategy evolution intensity threshold, response linkage measures will be implemented first, but resource reconstruction operations at the monitoring strategy level will not be carried out for the time being.

[0060] When the structural intervention urgency index is lower than the structural intervention urgency threshold and the monitoring strategy evolution intensity index is higher than the monitoring strategy evolution intensity threshold, the monitoring strategy evolution mechanism should be triggered to reschedule the response activation frequency.

[0061] When the structural intervention urgency index is higher than the structural intervention urgency threshold, and the monitoring strategy evolution intensity index is higher than the monitoring strategy evolution intensity threshold, the response linkage mechanism and the monitoring strategy reconstruction module are triggered simultaneously to carry out a coordinated closed-loop response adjustment of structural protection and resource scheduling.

[0062] The technical effects and advantages of the safety monitoring system for steel truss columns of elevated climbing scaffolding wall supports that integrates big data analysis as described in this invention are as follows:

[0063] This invention achieves efficient collection and analysis of multimodal response data of steel truss columns and wall-attached support areas during floor construction by constructing a multi-module collaborative intelligent monitoring system;

[0064] First, behavioral labels are extracted based on construction behavior control signals to construct a semantic mapping relationship between construction behavior and structural response, generating a behavior-driven dataset structure. Then, disturbance change trends are extracted through node response tensor and time difference analysis, and behavioral labels are fused to construct a structural state vector, enabling dynamic division and state evolution representation of the construction phase. Based on this, the system establishes a structural behavior distribution model, analyzes the deviation between the current state and the model response, and identifies the spatial propagation path of anomalies using a structural layout diagram, employing graph clustering to extract high-risk anomaly clusters. Finally, through temporal collaborative analysis of structural state changes and construction behavior in anomaly areas, differentiated monitoring and response strategies are triggered, enabling proactive identification and dynamic control of structural risks during construction, significantly improving the efficiency and accuracy of structural safety monitoring and response during the floor construction phase. Attached Figure Description

[0065] Figure 1 This is a structural schematic diagram of a safety monitoring system for steel truss columns of elevated climbing scaffolding with wall attachment, which integrates big data analysis according to the present invention. Detailed Implementation

[0066] 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.

[0067] In order to achieve the above objectives, Figure 1A structural schematic diagram of a safety monitoring system for steel truss columns of climbing scaffolding with wall attachment in an elevated floor, which integrates big data analysis, is provided. Specifically, it includes an elevated floor monitoring and data acquisition module, a structural state change module, a structural area anomaly identification module, and a response strategy module. The modules are connected by signals.

[0068] The elevated floor monitoring and data acquisition module is used to collect multimodal response information of steel truss columns and wall-attached support areas. It extracts behavior labels from the collected data according to the action type and state switching point, establishes a mapping relationship between construction behavior and structural response, and generates behavior-driven semantic structural response data.

[0069] The structural state change module is used to construct node response tensors based on structural response data, extract the perturbation change trend of structural nodes by combining time difference, and inject behavioral labels to generate node activation functions in the process of structural node state expression, thereby constructing the structural state vector of structural nodes.

[0070] The structural anomaly identification module is used to establish a structural behavior distribution model based on construction behavior conditions, identify the degree of anomaly deviation based on the response difference between the current state and the model, analyze the spatial propagation characteristics of anomalies based on the structural layout diagram, and extract anomaly cluster regions using graph clustering.

[0071] The response strategy module is used to acquire the structural state evolution characteristics within high-risk anomalous clumps in the anomalous clump region, analyze the temporal coordination information between structural response behavior and construction process, and trigger different monitoring response strategies based on the analysis results.

[0072] Step 1 involves collecting and uniformly modeling multi-source, multi-modal information on the climbing scaffold wall supports and steel truss column areas in the elevated floor construction scenario. This results in a dataset with time synchronization, semantic recognition capabilities, and behavioral response mapping relationships. The specific steps are as follows:

[0073] In the construction scenario, the steel truss columns in the elevated floor area serve as the main load-bearing components, and the wall-mounted supports act as the connection and stress-bearing nodes of the climbing scaffold system. Therefore, considering the structural characteristics of the elevated floor construction, the core monitoring units are determined to include:

[0074] The component response unit includes the upper and lower three sections of the steel truss column and its connection with the node; the connection response unit includes the rigid contact surface of the wall-mounted support and the possible sliding area; the construction behavior unit includes the lifting and lowering operation process of the climbing formwork and its control status.

[0075] Multiple types of sensors are deployed at key structural locations (the data acquisition units) to acquire multimodal response information. Specifically, strain gauges are installed at the top, middle, and bottom sections of the steel truss columns to acquire the strain response of the columns under vertical loads and horizontal disturbances, reflecting changes in the stress on the truss columns. Displacement sensors are installed between the wall supports and the structural walls to monitor micro-slippage, loosening, and relative displacement at the wall attachment points, identifying the slippage trend of the wall supports. High-frequency accelerometers are installed at node connections to capture rapid impact and vibration responses during the climbing formwork's lifting process. Behavior acquisition devices are connected to the climbing formwork's electrical control system to collect lifting execution commands, action start / end signals, and lifting displacement change data. A vision module can be optionally deployed to assist in the recognition of the climbing formwork's actions.

[0076] The above-mentioned equipment is used to acquire raw data covering all elements of structural deformation, local disturbance and construction behavior. The sampling frequency should be at least 50Hz to meet the timeliness requirements of data in dynamic construction scenarios, thereby forming a raw dataset. The raw dataset includes joint information on structural response and construction behavior.

[0077] Set up a unified global sampling time series, based on the least common multiple sampling step size, and resample or interpolate all data to ensure that all types of data have fully aligned structural response and behavioral state information at any point in time.

[0078] After parsing the collected data, the action type and state transition point are extracted to form behavior labels, which are used to indicate whether the current state is raised, steady state, unlocked or stopped.

[0079] A response mapping between construction actions and structural nodes is constructed. In actual construction, construction actions will not affect all structural nodes simultaneously. The scaffolding lifting process mainly affects the wall attachment points and the columns within a certain height range below them, while the unlocking action mainly affects the connection nodes and the adjacent structures above and below. A mapping relationship between construction actions and structural response units is established. The mapping relationship can be defined based on the structural layout drawing, construction equipment layout, and on-site experience. It is represented in the system in the form of a response influence diagram, i.e., the behavior state. Mapped to a set of structural response nodes For example: the act of raising the rack Mapped to Locking behavior Mapped to connection node ;

[0080] The time-synchronized acquired data, behavior labels, and structural response node sets (response mapping sets) are merged to form a unified structured data output unit. This unit contains the multimodal response values ​​(strain, displacement, vibration) of all controlled nodes at the current time point, the current behavior label (for model assessment), the active response region corresponding to the behavior (for local calculation optimization), and the node number and physical location (for structural mapping). The final data structure is as follows: ,in The structural response vector, For behavioral tags, For the set of structure nodes mapped to the current behavior, This is structural response data.

[0081] In summary, this step achieved unified acquisition of multimodal perception data, and also completed data time alignment, behavior label generation, and semantic mapping between construction behavior and structural response. Through this series of structured processing, the raw data was transformed into a dataset with structural recognition capabilities and behavior-driven logic.

[0082] Step two involves constructing structural state vectors and processing feature sequences. The structural response data is further organized into structural state vectors that possess temporal continuity, spatial comparability, behavioral relevance, and trend sensitivity. A feature evolution sequence reflecting the structural change process is then constructed. The specific steps are as follows:

[0083] Modal response integration and nodal state tensor generation are performed on structural response data. The monitoring system collects data including multiple modes, such as the strain response of steel truss columns, displacement changes between nodes, and vibration signals at connections. Various sensors reflect different response mechanisms of the structure. These are then integrated into a unified modal response tensor. Taking a specific monitoring node i as an example, at a certain time point... Construct its modal response tensor: ,For example, This represents the strain response at node i. This represents the axial displacement of node i. The frequency domain transformation of node i represents the vibration energy extracted from node i, and M represents the number of modes.

[0084] By encapsulating multiple dimensions of the structural response into a single state body, information loss or misinterpretation due to inconsistent dimensions can be avoided.

[0085] The elevated floor structure will experience large-scale disturbances during the climbing scaffolding process. Under construction disturbances, the structural response of the nodes exhibits a continuous change process, in which abrupt changes or nonlinear trends are often important signals for risk identification. The temporal differences of each structural response value in the modal response tensor are calculated to extract its velocity and acceleration over time, and then arbitrary modal dimensions are given. At any moment and the previous moment In between, construct its response mutation function: ,in, Indicates the time of the i-th node The structural response value of the m-th modal dimension at time m;

[0086] Trend feature sequences are extracted from the raw collected data, thereby enabling structural state evolution modeling over time.

[0087] Injecting behavioral labels into the state representation of structural nodes This enables the state vector to have the ability to interpret behavioral context. The behavioral state not only affects the structural response itself, but also determines whether the current node is in a controlled active region. For example, stress fluctuations generated during the climbing and raising of the scaffold may be a reasonable structural response; however, if similar fluctuations occur in the locked state, they should be considered abnormal.

[0088] When constructing the structure state vector, a behavior label needs to be introduced to drive the response of each node. The construction method is as follows:

[0089] Get every time Behavioral tags (e.g., lifting, steady state, lowering);

[0090] Find the set of structural response nodes corresponding to the behavior label. That is, the set of nodes that may be disturbed by this behavior;

[0091] If node Then let the activation coefficient be... ;

[0092] Otherwise, assume This indicates that the node should not have a significant response in its current behavior;

[0093] Construct node activation function The definition is as follows: ,in, This parameter represents the strength of the current node's influence relative to its behavioral state. It can be determined from historical disturbance statistics or field experience. This refers to the response impact area corresponding to this behavior;

[0094] This represents the intensity of the structural monitoring node's action under a specific construction behavior state. It is used to measure the node's sensitivity and participation in the structural response when that behavior occurs. The calculation expression is: T represents the number of time steps during which the construction activity continues. , These are the times of node i at time i. and The total structural response;

[0095] By combining the modal response tensor, trend perturbation function, and node activation function, a node at time t is constructed. Structure state vector: ;

[0096] in, For modal response tensors, This is a characteristic of trend disturbance. Each element represents the trend perturbation characteristics of structural node i in the corresponding modal dimension. Together, they constitute the physical and dynamic joint expression of the node. The activation weight ensures the relevance of its behavior and avoids irrelevant perturbations from entering the analysis path, so that in the construction scenario of the elevated floor, there is a coupling relationship between the climbing frame lifting behavior and the structural response.

[0097] Step 3: Conduct structural anomaly identification and risk assessment. The specific steps are as follows:

[0098] Structural response behavior is highly dependent on the construction stage. For example, during the lifting process, the strain and displacement of nodes may increase briefly, which is a normal disturbance. However, if the same response occurs in the steady-state locking stage, it may indicate loose connections or support slippage. Judging solely by the response amplitude is obviously unreliable. Therefore, it is necessary to establish statistical or graphical models of the normal structural state under different construction conditions.

[0099] The structural state vector sequence obtained from step two To identify structural anomalies, a structural behavior distribution model under normal operating conditions is established for subsequent deviation assessment. Since the structural response is strongly dependent on the construction behavior state, a structural state condition distribution model under construction behavior conditions should be constructed, defining the state distribution function as follows: ,in, Indicates the time of the i-th structural node. The state vector, Indicates the moment Corresponding construction behavior tags, For a specific construction activity (such as scaffolding) steady state wait); Indicates that the construction activity is in a state of... The state vector of node i at state i The conditional probability distribution, in other words, is given the construction state as... Under the premise that the node response state may occur, the probability distribution of the node response state is as follows;

[0100] Real-time deviation detection and anomaly intensity identification of state vectors are performed to determine the degree of deviation. For each structural node, at the current moment, the corresponding conditional distribution model is activated based on the construction label, the typical response trajectory of the node under this behavior is extracted from the model, and the deviation intensity score between the current state vector and the reference vector is calculated. ,in, This represents the actual state response value of the m-th modal dimension of the state vector; Indicates behavioral conditions The reference trajectory value for the m-th modal dimension. For deviation mapping function;

[0101] Actual state response value Through real-time sensors The state representation is obtained after constantly collecting data and standardizing and mapping it. For example, raw multimodal response data is collected, where the m-th mode may correspond to the first dimension of strain, the second dimension of displacement, the third dimension of vibration energy, etc. Multiple sensors configured at node i (such as strain gauges, displacement sensors, and accelerometers) will collect different types of response data in real time. All modal data are aligned according to a unified timestamp and transformed into state values ​​that express the dynamic changes of the structure through response processing functions (such as rate of change, slope, and local partial derivatives). The m-th dimension response value is then transformed from the state vector. The solution is: ;

[0102] Reference trajectory value In the construction activity In this scenario, the "normal state" of the m-th modal dimension, obtained through historical structural state trajectories or modeling estimation, is used as a reference baseline for deviation judgment; specifically, historical data is obtained by categorizing it according to behavioral labels. (Such as lifting and locking) are extracted in segments, that is, all The state vector data is grouped, and the m-th modal dimension data of node i is modeled for all time periods. Methods include time window mean / median extraction (for simple scenarios), distribution fitting (such as Gaussian distribution, kernel density estimation), etc. Given the relative time position corresponding to a given moment, the expected response under the current behavioral state is extracted from the reference model. .

[0103] A specific definition of the deviation mapping function is as follows: Where 'a' represents the current time. The actual state response value of the lower structure node i in the m-th modal dimension, such as, ; b indicates the conditions of construction behavior The reference trajectory value of the next node i in the m-th modal dimension is as follows: The use of asymmetric enhancement function processing can avoid misjudgment due to amplification of small perturbations, while highlighting the expression of large nonlinear anomalies.

[0104] Deviation strength score is an indicator of the anomalous significance of the node in its current behavioral state.

[0105] Furthermore, a Markov latent variable graphical model is used for modeling, that is, to build a state transition graph for each node; a latent variable space is established for the response trend of each modal dimension, and the node state trajectory transition probability is inferred under each behavior label condition.

[0106] Abnormal signals have spatial propagation characteristics, especially in the area where the climbing formwork wall support is connected to the steel truss, where minute disturbances may be rapidly propagated through the structural coupling path;

[0107] Perform spatial clustering analysis and propagation trend identification of structural anomalies, that is, perform spatial clustering analysis on the deviation values ​​of all nodes to identify local anomaly clustering areas and propagation paths;

[0108] The structural layout diagram is transformed into a structural atlas G=(N,E), where nodes are monitoring points (structural response points), edges are component connection relationships, edge weights are assigned to represent the coupling strength between nodes, and an anomaly propagation model is run on the graph to evaluate whether the current anomaly has spatial consistency or whether it spreads from a certain point.

[0109] Define the degree of abnormal deviation on the structural layout drawing: ,in , These represent the deviation strength scores for nodes i and j, respectively. The weight of the connection edge between the i-th node and the j-th node in the structural layout diagram can be set according to actual needs, usually based on information such as structural layout or material stiffness; E represents the set of edges in the structural layout diagram, which is the set of connection relationships between each monitoring node in the structure (e.g., the connection between steel truss columns and wall supports). This indicates that node i and node j have a direct connection or a structural coupling relationship in the structure.

[0110] The degree of abnormal deviation is used to assess the non-cooperation of structural response anomalies along the connection path. A high value indicates that the anomaly may have started propagating from a certain point; a low value indicates that the anomaly is localized. A concentrated upward movement suggests the possible existence of local structural instability or slippage.

[0111] By combining graph clustering algorithms (such as spectral clustering) to extract abnormal clusters in high-deviation regions, the specific affected areas can be identified. The specific steps are as follows:

[0112] The system will establish a spatial connection diagram of the entire structural monitoring area based on structural drawings, component layout information, and sensor deployment locations. In this structural diagram, each monitoring node is regarded as a structural unit, and the physical components connecting these units (such as columns, beams, supports, etc.) are used as edges between nodes, forming a network diagram that reflects the spatial layout of the structure and the load transfer path.

[0113] The abnormal deviation degree of each node is calculated in real time. The deviation degree is used to measure the difference between the current structural response state (such as strain, displacement, vibration, etc.) of the node and its normal response range under the current construction behavior background.

[0114] The system will integrate the abnormal deviation of these nodes into the structural connection diagram to form a composite structural diagram with structural connection relationships and abnormal performance levels. This means that each node in the diagram has its current abnormality score, and the lines between nodes represent physical connection relationships and response coupling capabilities.

[0115] By combining the structural connection strength with the abnormal deviation differences between each node, an abnormal similarity map is constructed and similarity analysis is performed.

[0116] In this graph, the strength of the connection between adjacent nodes depends not only on whether the nodes are physically connected, but also on whether the abnormal behavior of the nodes is consistent. If two nodes are in a high deviation state and the degree of abnormality between the nodes is very similar, then the connection between the nodes will be judged by the system as a high abnormality similarity connection. Thus, the system can not only identify which nodes are abnormal themselves, but also identify which abnormal nodes constitute a consistent regional pattern.

[0117] Graph clustering technology is used to divide the entire anomaly similarity map into regions, automatically identifying groups of nodes that are physically adjacent to each other and highly consistent in their anomaly behavior. Each identified cluster region can be regarded as a potential structural risk area or an area affected by disturbance.

[0118] After clustering is completed, the system will count the degree of abnormal deviation of all nodes in each cluster region and calculate the overall risk level of the region. If the degree of abnormality of nodes in a certain region is generally high, or the abnormal value of nodes in the region shows an upward trend, the system will mark the region as a high-risk abnormal cluster.

[0119] Step four involves conducting hierarchical response strategy linkage and monitoring strategy self-evolution analysis. This involves transforming the results of structural anomaly identification and risk assessment into control strategies with hierarchical response capabilities and dynamic adjustability. Furthermore, through a risk feedback mechanism, the monitoring strategy is adaptively optimized and the system is updated through evolution. The specific steps are as follows:

[0120] Obtain the structural state evolution characteristics within high-risk abnormal cluster areas, and analyze the temporal coordination information between structural response behavior and construction process. The temporal coordination information includes the structural intervention urgency index and the monitoring strategy evolution intensity index.

[0121] The Structural Intervention Urgency Index is used to indicate whether a structure is in a high-risk state that requires immediate intervention at the current moment. The Structural Intervention Urgency Index measures the degree of abrupt change in structural disturbance, the lag in response recovery, and the temporal correlation between disturbance and construction behavior. It is used to determine whether to trigger a coordinated response strategy, such as suspending construction or locking key structural nodes.

[0122] The structural intervention urgency index supports the system's judgment of response strategies for abnormal areas. If the structural intervention urgency index value rises significantly, it indicates that the structural disturbance is severe, the recovery is slow, and it is closely related to construction actions. The system should immediately issue control commands, such as suspending the climbing formwork lifting, locking the wall attachments, or issuing local / remote alarms. It is a key decision signal connecting the structural monitoring and judgment module and the engineering linkage control module, realizing a closed-loop drive from state perception to engineering intervention.

[0123] The logic for obtaining the structural intervention urgency index is as follows:

[0124] Obtain the time when the disturbance response reaches a local maximum when the disturbance occurs. Obtain the structural response to recover to the stable threshold. The first time point Calculate the disturbance recovery time interval: ;

[0125] Obtain the backtracking window length T, take T sampling points backward from the current time, and calculate the first derivative sequence of the curve. , used to represent the rate of change of modal response with time, r represents the sequence of modal response values, t is the time of the sampling point, each time the derivative changes from positive to negative or from negative to positive, it is denoted as a mutation, and the perturbation mutation density is calculated: ,in, This is an indicator function; the value is 1 if the sign changes, and 0 otherwise.

[0126] Record the set of time points for all construction behavior state transitions and identify the time from the peak disturbance. Recent Switching Point Calculate the peak time interval: The formula for calculating the structural intervention urgency index is: .

[0127] It should be noted that the structure refers to the steel truss column structure of the climbing scaffold wall support in the elevated floor. The stability threshold is set according to the safety requirements of the actual scenario. After the structure is disturbed, the response curve starts to decline from the peak. The system detects the decline in response value and stabilizes at the time point. The time difference is the structure recovery period. The larger the time, the more sluggish the structure is. If the disturbance peak is very close to the behavior change time, this value will be large, indicating that the disturbance may be caused by construction behavior and should be more inclined to respond. r represents the modal response value sequence, that is, the response curve of a certain structural node under a certain mode. For example, if it is a strain mode, then r represents the strain value; if it is a displacement mode, then r represents the displacement value; if it is a vibration mode, then r represents the vibration energy, etc.

[0128] The monitoring strategy evolution intensity index is used to indicate whether the current monitoring system strategy needs to be adaptively adjusted. The monitoring strategy evolution intensity index measures the complexity of the structural state, the instability of the response trend and the integrity of the sampled data. It is used to determine whether it is necessary to adjust the sampling frequency, reconstruct the sampling node priority or optimize the monitoring resource allocation strategy.

[0129] The role of the monitoring strategy evolution intensity index is to drive the strategy adaptive mechanism of the monitoring system. When the monitoring strategy evolution intensity index increases, it indicates that the structural state is becoming more complex and the data pressure is increasing. The system will actively compress the sampling tasks of low-risk nodes and increase the monitoring density of high-risk areas. If necessary, it can also update the model input dimension or enable dynamic feature channels. The monitoring strategy evolution intensity index ensures that the system still has the ability to operate continuously with high response, high resolution and high resource utilization in scenarios with limited resources and drastic structural evolution.

[0130] The logic for obtaining the monitoring strategy evolution intensity index is as follows:

[0131] Obtain the state vector of the i-th node of the structure. and the centroid of all state vectors The envelope volume formed by this set of state vectors in the manifold space. , This represents the state vector of all structural nodes at the current moment. The structural response manifold complexity of the response manifold spanned in space is expressed as: In the formula, N is the total number of nodes in the structure;

[0132] Obtain the total number of data points that should be sampled for the i-th node in the current time window, as well as the number of data entries that failed sampling, were missing, or had errors in the return data. Calculate the sampling completeness using the following expression: , This represents the total number of data points that should be sampled for the i-th node in the current time window. This represents the number of data entries that failed to be sampled at the i-th node within the time window. The evolution intensity index of the monitoring strategy is calculated using the following expression: .

[0133] It should be noted that the centroid of the state vector refers to the center vector, and the structural response manifold complexity reflects the degree of diffusion of state points of multiple nodes in high-dimensional space. The larger the coverage volume and the higher the dispersion, the more complex the structural state, and the more the system needs to increase the sampling frequency.

[0134] A structural intervention urgency threshold is set to determine whether the structure is currently in a critical state that requires triggering an intervention response. A monitoring strategy evolution intensity threshold is set to determine whether the current monitoring strategy is still applicable to the trend of structural state evolution. The thresholds are compared and analyzed with the structural intervention urgency index and the monitoring strategy evolution intensity index, respectively.

[0135] When the structural intervention urgency index is lower than the structural intervention urgency threshold and the monitoring strategy evolution intensity index is lower than the monitoring strategy evolution intensity threshold, it indicates that the current disturbance waveform of the structure is relatively stable, and the response after the disturbance can fall back to the stable range in a short time. This indicates that the structure has not shown any continuous abrupt change or signs of instability, and the disturbance effect of construction behavior on the structural state is within a controllable range. At the same time, the structural response vector in the state space maintains a low-dimensional convergence trend, and there is no frequent reversal of the response direction or divergence of the characteristic trajectory. The sampled data has high temporal integrity and spatial coverage. The system can determine that the structure is in an acceptable risk range and maintain the current observation strategy and response quiescent state.

[0136] When the structural intervention urgency index is higher than the structural intervention urgency threshold and the monitoring strategy evolution intensity index is lower than the monitoring strategy evolution intensity threshold, it indicates that the structure has a significant area of ​​dense disturbance reversal within a local time window, and there is a lag in the response recovery after the disturbance peak. At the same time, the time distance between this disturbance event and the most recent construction behavior switch is short, indicating that the current structural response has become highly sensitive to external behavior and has construction-induced characteristics. The system still maintains good information continuity between sampling nodes, the expansion degree of the state manifold in space is limited, and there is no continuous interruption or trend jump in the sampling data. Based on this, the system determines that response linkage measures should be implemented first to prevent the expansion of structural anomalies, but resource reconstruction operations at the monitoring strategy level will not be performed for the time being.

[0137] When the structural intervention urgency index is lower than the structural intervention urgency threshold and the monitoring strategy evolution intensity index is higher than the monitoring strategy evolution intensity threshold, it indicates that the current disturbance response of the structure has a rapid decline characteristic, the disturbance amplitude can be controlled within the steady state range in a short period of time, and the disturbance waveform does not show high-frequency reversal behavior in the local derivative density distribution. The response time interval between the structure and the construction behavior is in the non-sensitive range, and the risk inducement is weak. At the same time, the multi-node response status shows a local non-consistent distribution in the embedded space, the characteristic trajectory direction is scattered, and some nodes have intermittent sampling information and sensing data coverage gaps. This indicates that there is a matching deviation between the current monitoring strategy and the structural state evolution. The system should trigger the monitoring strategy evolution mechanism accordingly to reschedule the response activation frequency to ensure the accuracy of state perception.

[0138] When the structural intervention urgency index is higher than the structural intervention urgency threshold, and the monitoring strategy evolution intensity index is higher than the monitoring strategy evolution intensity threshold, it indicates that the structure has entered a high-risk state range where enhanced disturbance and delayed response coexist. The disturbance curve exhibits high-frequency nonlinear jump characteristics within the near-time window, the response hysteresis interval is significantly prolonged, and the disturbance peak coincides highly with the construction behavior switching time, exhibiting a typical behavior-induced unsteady-state response mode. At the same time, the structural state vector shows a high-dimensional expansion trend in the embedded manifold space, the differences in feature expression between nodes are significantly enhanced, some sampling points experience transmission interruption or feedback delay, and the system's state perception capability is interfered with. The response linkage mechanism and the monitoring strategy reconstruction module should be triggered simultaneously to carry out coordinated closed-loop response adjustment and optimization of structural protection and resource scheduling.

[0139] Based on the threshold comparison results, the current risk level is determined (e.g., low risk, localized concentrated risk, dispersed and diffused risk, global high risk, etc.), and this level is mapped to the structural space to generate structural partition identifiers with risk level labels. According to the determined risk level, the corresponding set of response strategies is matched, including whether to trigger early warning prompts, whether to initiate local re-inspection or modal recalibration, whether to require suspension of construction or loading processes in a certain area, and whether to enter the full structure emergency response state. At the same time, the system needs to convert the selected strategies into executable plans.

[0140] In summary, this invention achieves efficient collection and analysis of multimodal response data of steel truss columns and wall-mounted support areas during floor construction by constructing a multi-module collaborative intelligent monitoring system.

[0141] First, behavioral labels are extracted based on construction behavior control signals to construct a semantic mapping relationship between construction behavior and structural response, generating a behavior-driven dataset structure. Then, disturbance change trends are extracted through node response tensor and time difference analysis, and behavioral labels are fused to construct a structural state vector, enabling dynamic division and state evolution representation of the construction phase. Based on this, the system establishes a structural behavior distribution model, analyzes the deviation between the current state and the model response, and identifies the spatial propagation path of anomalies using a structural layout diagram, employing graph clustering to extract high-risk anomaly clusters. Finally, through temporal collaborative analysis of structural state changes and construction behavior in anomaly areas, differentiated monitoring and response strategies are triggered, enabling proactive identification and dynamic control of structural risks during construction, significantly improving the efficiency and accuracy of structural safety monitoring and response during the floor construction phase.

[0142] The above formulas are all dimensionless calculations. The formulas are derived from software simulations using a large amount of collected data, and are the closest to the real situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0143] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0144] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0145] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0146] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0147] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. 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 safety monitoring system for steel truss columns supporting wall-mounted scaffolding in elevated floors, integrating big data analysis, characterized in that: It includes a data acquisition module for monitoring the elevated floor, a structural state change module, a structural area anomaly identification module, and a response strategy module. The modules are connected by signals. The elevated floor monitoring and data acquisition module is used to collect multimodal response information of steel truss columns and wall-attached support areas. It extracts behavior labels from the collected data according to the action type and state switching point, establishes a mapping relationship between construction behavior and structural response, and generates structural response data with behavior-driven semantics. The structural state change module is used to construct node response tensors based on structural response data, extract the perturbation change trend of structural nodes by combining time difference, and inject behavioral labels to generate node activation functions in the process of structural node state expression, thereby constructing the structural state vector of structural nodes. The structural anomaly identification module is used to establish a structural behavior distribution model based on construction behavior conditions, identify the degree of anomaly deviation based on the response difference between the current state and the model, analyze the spatial propagation characteristics of anomalies based on the structural layout diagram, and extract anomaly cluster regions using graph clustering. The response strategy module is used to acquire the structural state evolution characteristics within high-risk anomalous clumps in the anomalous clump region, analyze the temporal coordination information between structural response behavior and construction process, and trigger different monitoring response strategies based on the analysis results. Behavioral tags include whether the vehicle is currently in a raised, steady-state, unlocked, or stopped state; Furthermore, behavioral labels are injected into the structural node state representation process to generate node activation functions, thereby constructing the structural node's structural state vector. Specific steps include: Get every time Behavioral tags ; Find the influence structure region corresponding to the behavior label Obtain the set of nodes whose behavior causes disturbance; If node Then set the node activation function as Otherwise, set the node activation function. ,in, This is the effect strength parameter of the current node relative to the behavioral state. The area affected by the response to the behavior; Where T represents the number of time steps in which the construction activity continues. , These are the times of node i at time i. and The total structural response; By combining the modal response tensor, trend perturbation function, and node activation function, a node at time t is constructed. Structure state vector: ,in, For modal response tensors, This is the trend disturbance function.

2. The safety monitoring system for steel truss columns of elevated climbing scaffold wall supports integrating big data analysis as described in claim 1, characterized in that: This method is used to collect multimodal response information of steel truss columns and wall-attached support areas. Behavioral labels are extracted from the collected data based on action type and state transition points. The specific steps are as follows: Multimodal response information of steel truss columns and wall-mounted supports is collected. The monitoring and collection units include the upper and lower sections of the steel truss columns and the node connections, the rigid contact surface and sliding area of ​​the wall-mounted supports, and the lifting and lowering operation process of the climbing frame. Sensors are deployed in the data acquisition unit, including strain gauges placed in the upper, middle and lower sections of the steel truss column, to obtain the strain response of the column under vertical load and horizontal disturbance. Displacement sensors are installed between the wall support and the structural wall to monitor micro-slippage, loosening, and relative displacement at the wall attachment points. High-frequency accelerometers are installed at the node connections to capture rapid impact and vibration responses during the climbing frame's lifting process; The behavior acquisition device is connected to the climbing scaffold electrical control system to collect lifting execution commands, action start and end signals, and lifting displacement change data. Set a unified global sampling time series, and resample or interpolate all collected data based on the least common multiple sampling step size; The collected data is analyzed to extract action types and state transition points, forming behavior labels.

3. The safety monitoring system for steel truss columns supporting climbing scaffolding in elevated floors, integrating big data analysis, as described in claim 2, is characterized in that: Establishing a mapping relationship between construction behavior and structural response, and generating structural response data with behavior-driven semantics, includes the following specific steps: Establish a mapping relationship between construction behavior and structural response units. The mapping relationship is determined based on the structural layout diagram, construction equipment layout and on-site experience. It is represented in the form of a response influence diagram in the system, which maps the behavior state to a set of structural response nodes. The time-synchronized collected data, behavioral tags, and structural response node sets are merged to form structural response data; The structural response data includes the modal response values ​​of all controlled nodes at the current time point, the current behavior label, the active response area corresponding to the behavior, the node number and physical location data.

4. The safety monitoring system for steel truss columns of elevated climbing scaffold wall supports integrating big data analysis as described in claim 3, characterized in that: This method is used to construct node response tensors based on structural response data, extract the perturbation change trends of structural nodes by combining time difference, and inject behavioral labels into the structural node state representation process to generate node activation functions, thereby constructing the structural node's structural state vector. The specific steps are as follows: The collected data are integrated into a modal response tensor. The time difference of each response value in the modal response tensor is calculated, the velocity and acceleration that change with time are extracted, and the modal dimension of adjacent time points is given to determine the response mutation function. Trend feature sequences are extracted from the raw collected data, and structural state evolution modeling is performed based on time. A behavior label is introduced to drive the response of each node, and the structural state vector of the node is constructed.

5. A safety monitoring system for steel truss columns supporting climbing scaffolding in elevated floors, integrating big data analysis, as described in claim 4, is characterized in that: This is used to establish a structural behavior distribution model based on construction behavior conditions, and to identify the degree of abnormal deviation based on the response difference between the current state and the model. The specific steps are as follows: Establish a structural behavior distribution model under normal operating conditions, perform deviation judgment, and set the state distribution function as follows: ,in, Indicates the time of the i-th structural node. The state vector, Indicates time Corresponding behavioral tags For specific construction activities; Indicates that the construction activity is in a state of... The state vector of node i at state i The conditional probability distribution; Real-time deviation detection and abnormal deviation identification of state vectors are performed. For each structural node i, at the current moment, the typical response trajectory of the node under this behavior is extracted based on the behavior label, and the deviation strength score between the current state vector and the reference vector is calculated. ,in, This represents the m-th modal dimension of the state vector; Indicates behavioral conditions The reference trajectory value in the m-th dimension. For deviation mapping function; Actual state response value Through real-time sensors The reference trajectory value is obtained by collecting data at all times and then standardizing and mapping the state representation. In the construction activity In this scenario, the normal state value of the m-th modal dimension obtained through the historical structural state trajectory is used as the baseline for deviation judgment; the deviation intensity score is the abnormal significance index of the node in the current behavioral state. Modeling is performed using a Markov latent variable graphical model to establish a state transition graph for each node and to establish a latent variable space for the response trend of each modality dimension. Under each behavior label condition, the probability of node state trajectory transition is inferred.

6. The safety monitoring system for steel truss columns of elevated climbing scaffold wall supports integrating big data analysis as described in claim 5, characterized in that: Based on the analysis of the spatial propagation characteristics of anomalies using structural layout diagrams, graph clustering is used to extract anomaly cluster regions. The specific steps are as follows: Perform spatial clustering analysis and propagation trend identification of structural anomalies, that is, perform spatial clustering analysis on the deviation values ​​of all nodes to identify local anomaly clustering areas and propagation paths; The structural layout diagram is converted into a structural atlas, where nodes represent monitoring points and edges represent component connections. Define the degree of abnormal deviation on the structural layout drawing: ,in These represent the deviation strength scores for nodes i and j, respectively. This represents the weight of the connection edge between the i-th node and the j-th node in the structural layout diagram; , indicates that node i and node j have a direct connection or structural coupling relationship in the structure; E represents the set of edges in the structural layout diagram; By combining graph clustering algorithms, abnormal clusters are extracted from the off-center areas, the affected areas are identified, and high-risk abnormal clusters are determined.

7. A safety monitoring system for steel truss columns supporting climbing scaffolding in elevated floors, incorporating big data analysis, as described in claim 6, is characterized in that: By combining graph clustering algorithms to extract anomalous clusters from off-center regions, affected areas are identified, and high-risk anomalous clusters are determined. The specific steps are as follows: Real-time calculation of the degree of abnormal deviation of each node in the structural layout diagram; The degree of abnormal deviation of nodes is integrated into the structural connection diagram to form a composite structural diagram that has both structural connection relationships and degree of abnormal behavior. By combining the structural connection strength with the abnormal deviation differences between each node, an anomaly similarity map is constructed and similarity analysis is performed; Graph clustering techniques are used to partition the entire anomaly similarity graph, identifying sets of nodes that are physically adjacent to each other and highly consistent in their anomaly behavior. After clustering is completed, the degree of abnormal deviation of all nodes in each cluster region is calculated. If the degree of abnormal deviation of nodes in a region shows an upward trend, the region is marked as a high-risk abnormal cluster.

8. A safety monitoring system for steel truss columns supporting climbing scaffolding in elevated floors, incorporating big data analysis, as described in claim 7, is characterized in that: This method is used to acquire the structural state evolution characteristics within high-risk anomalous clumps in anomaly clump regions, analyze the temporal coordination information between structural response behavior and the construction process, and trigger different monitoring and response strategies based on the analysis results. The specific steps are as follows: Obtain the structural state evolution characteristics within high-risk anomalous cluster regions and analyze the temporal coordination information between structural response behavior and construction process; Temporal synergy information includes the structural intervention urgency index and the monitoring strategy evolution intensity index; The structural intervention urgency index is used to represent the abruptness of structural disturbances, the hysteresis of response recovery, and the temporal correlation between disturbances and construction activities. The monitoring strategy evolution intensity index is used to indicate the urgency of whether the current monitoring system strategy needs adaptive adjustment; Set thresholds for the intensity of monitoring strategy evolution and the urgency of structural intervention; The threshold for the intensity of monitoring strategy evolution and the threshold for the urgency of structural intervention were compared and analyzed with the intensity index of monitoring strategy evolution and the urgency index of structural intervention, respectively.

9. A safety monitoring system for steel truss columns supporting climbing scaffolding in elevated floors, incorporating big data analysis, as described in claim 8, is characterized in that: The monitoring strategy evolution intensity threshold and structural intervention urgency threshold are compared and analyzed with the monitoring strategy evolution intensity index and structural intervention urgency index, respectively. The specific steps are as follows: When the structural intervention urgency index is lower than the structural intervention urgency threshold and the monitoring strategy evolution intensity index is lower than the monitoring strategy evolution intensity threshold, the structure is in an acceptable risk range, and the current observation strategy and response remain silent. When the structural intervention urgency index is higher than the structural intervention urgency threshold and the monitoring strategy evolution intensity index is lower than the monitoring strategy evolution intensity threshold, response linkage measures will be implemented first, but resource reconstruction operations at the monitoring strategy level will not be carried out for the time being. When the structural intervention urgency index is lower than the structural intervention urgency threshold and the monitoring strategy evolution intensity index is higher than the monitoring strategy evolution intensity threshold, the monitoring strategy evolution mechanism should be triggered to reschedule the response activation frequency. When the structural intervention urgency index is higher than the structural intervention urgency threshold, and the monitoring strategy evolution intensity index is higher than the monitoring strategy evolution intensity threshold, the response linkage mechanism and the monitoring strategy reconstruction module are triggered simultaneously to carry out a coordinated closed-loop response adjustment of structural protection and resource scheduling.

Images