Steel structure grid high-altitude bulk construction method based on triangular rods

Abstract

The invention relates to the field of high-altitude construction, and discloses a steel structure grid high-altitude bulk construction method based on a triangular rod, which comprises the following steps of: synchronously hoisting and positioning a triangular rod component to be assembled and a corresponding node ball piece in a multi-hoisting-point cooperative high-altitude operation area to form a continuous traceable assembly posture change sequence; performing state tracking of a cross-node connection dependency chain on the assembly attitude change sequence, and constructing a multi-dimensional structure feature vector reflecting a construction sequence and a spatial constraint relationship; a splicing sequence evolution curve is generated through the multi-dimensional structure feature vectors, and according to the deviation degree of the splicing sequence evolution curve and the actual component in-place time sequence; dividing a local grid area where the dislocation risk node is located and a stress conduction range into a to-be-intervened operation area, and establishing a local construction relation graph with real-time updating capability; and implementing a dynamic intervention decision on the construction sequence evolution curve based on the local construction relation graph. The method has the advantage of improving the construction efficiency.

Description

Technical Field

[0001] This invention relates to the field of high-altitude construction, specifically to a method for high-altitude bulk construction of steel space frames based on triangular rods. Background Technology

[0002] Steel space frames are widely used in large-span buildings such as stadiums and exhibition centers. High-altitude assembly methods are widely adopted due to their ability to reduce ground assembly space requirements and shorten construction time. Existing high-altitude assembly methods for steel space frames based on triangular braces largely rely on manual labor to hoist individual members one by one from high-altitude platforms or suspended scaffolding and assemble them into triangular units. However, during construction, because the space frame members have relatively small cross-sections and long lengths, they are prone to slight swaying due to wind loads and member self-vibration during hoisting to high altitudes. This can lead to hole position deviations when workers connect bolt holes. Although such deviations are generally only 2–5 mm, in specific scenarios, such as the edge areas of large-span dome structures, the connection nodes of space frame members are often at non-standard angles. Any slight deviation can amplify the cumulative error of subsequent units, making overall assembly difficult and even requiring rework. Current technology lacks a method to effectively suppress member micro-swaying and quickly and accurately calibrate hole positions during the hoisting and positioning phase of high-altitude assembly, thus affecting construction efficiency and structural installation accuracy. Therefore, it is necessary to design a high-altitude bulk construction method for steel space frames based on triangular rods to improve construction efficiency. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a method for high-altitude bulk construction of steel space frames based on triangular rods, which has the advantage of improving construction efficiency and solves the problems mentioned in the background technology.

[0004] To achieve the aforementioned goal of improving construction efficiency, this invention provides the following technical solution: a method for high-altitude bulk assembly construction of steel space frames based on triangular rods, comprising the following steps: In the high-altitude operation area with multi-point cooperation, the triangular rod components to be assembled and the corresponding node ball components are lifted and positioned synchronously. The attitude sensing measurement component is used to extract key spatial parameters reflecting the stability of the components. The measurement results are dynamically corrected by combining the sling force adjustment mechanism and the high-altitude displacement compensation unit to form a continuous and traceable assembly attitude change sequence. State tracking of cross-node connection dependency chains is performed on the assembly posture change sequence. Combining historical installation cycle records, success rate of adjacent node docking, and component vibration attenuation law, a multi-dimensional structural feature vector reflecting the construction sequence and spatial constraint relationship is constructed. The assembly sequence evolution curve is generated from the multidimensional structural feature vector. Based on the degree of deviation between the assembly sequence evolution curve and the actual component arrival time sequence, the node positions with installation misalignment risk are identified, and the assembly links corresponding to the node positions are prioritized. The local space frame area and stress transmission range where the installation misalignment risk node is located are divided into the intervention work area. The layout of workers, the stress state of the slings and the installation completion of adjacent components in the intervention work area are integrated to establish a local construction relationship map with real-time updating capability. Based on the local construction relationship map, combined with the priority calibration results and the hoisting micro-sway characteristics presented in the assembly posture change sequence, dynamic intervention decision is made on the construction sequence evolution curve.

[0005] Preferably, the process of extracting key spatial parameters reflecting the stability of a component using an attitude sensing measurement component is as follows: By deploying three-dimensional attitude sensors and angular velocity sensors around the suspension points and nodal spheres, the pitch angle, deflection angle and axial tilt of the triangular rod component in the high-altitude environment are collected in real time. The parameters are then used as the spatial parameters of the component after time synchronization and filtering. By combining the wind load on the component during the hoisting process with the stress distribution fed back by the sling force sensor, the attitude offset is calculated and added to the spatial parameters to form a complete attitude measurement result; The component design reference values ​​are obtained based on the structural drawings and finite element simulation calculation results during the construction design stage, including the ideal posture angle and force balance state of the component in the theoretical installation position; By comparing spatial parameters with component design benchmark values, key spatial parameters reflecting the components during the assembly process can be extracted.

[0006] Preferably, the process of forming a continuous and traceable sequence of assembly posture changes is as follows: Based on the key spatial parameters output by the attitude sensing measurement component, a time-sequential record of attitude changes is constructed. The system combines the sling force adjustment mechanism to correct instantaneous attitude deviations caused by uneven sling points or wind interference in real time. A high-altitude displacement compensation unit is introduced to dynamically correct abnormal fluctuation points in the recorded sequence, forming an assembly attitude change sequence.

[0007] Preferably, the process of performing cross-node connection dependency chain state tracking on the assembly attitude change sequence is as follows: The attitude parameters of key nodes in the assembly attitude change sequence are mapped to the node connection dependency chain, and the temporal correlation between each node is marked. By combining historical installation time records, the actual completion rate of the installation sequence between nodes is compared. The success and failure rates of docking between adjacent nodes are statistically analyzed. Combined with the vibration attenuation law of components during the assembly process, the evolution path of cross-node dependency chains is dynamically tracked, and the state trajectory that reflects both temporal and physical constraints is output.

[0008] Preferably, the process of constructing a multi-dimensional structural feature vector reflecting the construction sequence and spatial constraints is as follows: Map the attitude of each key node, docking result and component stress state in the cross-node dependency chain state trajectory to the feature space; By integrating the operational rhythm of workers, the force distribution of lifting points and the spatial position constraints of components during the construction process, a multi-dimensional set of statistical parameters is formed. By processing the parameter set through normalization and feature reduction algorithms, a multi-dimensional structural feature vector that can reflect the rationality of the construction sequence and the degree of spatial constraint tension is generated.

[0009] Preferably, the process for identifying node locations at risk of installation misalignment is as follows: Based on multi-dimensional structural feature vectors, an assembly sequence evolution curve is generated to extract the change pattern of the construction sequence. The assembly sequence evolution curve is compared with the actual component arrival time series to calculate the offset index; When the offset exceeds the preset threshold, the location of the critical node with potential installation misalignment risk is located by combining abnormal fluctuations in node attitude and docking delay signals.

[0010] Preferably, the process of prioritizing the assembly links corresponding to the node locations is as follows: The identified installation misalignment risk nodes and their upstream and downstream dependency chains will be the key monitoring targets; Based on the construction safety level, component span size, and stress transmission capacity of nodes, the assembly links are graded and evaluated, and different priority labels are assigned to the assembly links according to the evaluation results.

[0011] Preferably, the process of dividing the local grid area where the installation misalignment risk node is located and the stress transmission range into the intervention work area is as follows: Based on the spatial coordinates of the nodes at risk of misalignment during installation, the local grid area where the coordinates are located is determined. Based on structural mechanics analysis, calculate the stress transmission range corresponding to the misaligned node; The spatial extent of the local grid area is integrated with the stress-affected area to form the boundary of the area to be intervened.

[0012] Preferably, the process of establishing a local construction relationship map with real-time updating capability is as follows: Extract key elements from the work area to be intervened, including the arrangement of workers, the stress state of the slings, and the completion of the installation of adjacent components; Based on the actual construction progress and work scheduling plan, establish logical connections and physical constraints between nodes; By updating the relationship map through real-time monitoring data, the constantly changing working environment and structural status during construction are reflected, forming a local construction relationship map.

[0013] Preferably, the process of making dynamic intervention decisions on the construction sequence evolution curve is as follows: By combining the priority marking results in the local construction relationship diagram with the hoisting micro-swing characteristics in the attitude change sequence, the rationality of the current construction sequence is comprehensively evaluated. When a potential risk of misalignment is detected in the construction sequence, the installation order is adjusted based on the risk level, and the scheduling of workers and the distribution of force on the slings are optimized.

[0014] Compared with existing technologies, this invention provides a method for high-altitude bulk construction of steel space frames based on triangular rods, which has the following beneficial effects: This invention establishes a complete closed loop for high-altitude bulk construction, from attitude measurement to construction sequence intervention, thereby achieving precise control and risk management of multi-point coordination during the construction of steel structure space frames. By utilizing attitude sensing measurement components and sling force adjustment mechanisms, key spatial parameters of triangular rod components can be acquired and corrected in real time, ensuring stable posture of the components during lifting, movement, and positioning, reducing the risk of high-altitude deviation and vibration. Through cross-node connection dependency chain state tracking and multi-dimensional structural feature vector construction, quantitative analysis of the construction sequence and spatial constraint relationship can be achieved, accurately identifying nodes with potential installation misalignment risks. By combining the deviation analysis of the assembly sequence evolution curve and the actual arrival time sequence, priority can be assigned to high-risk nodes and their links, ensuring the safety and reliability of the critical path of construction. By dividing the work area to be intervened and establishing a local construction relationship map with real-time updating capability, construction managers can comprehensively grasp information such as the layout of workers, the stress state of slings, and the completion of component installation, enabling dynamic monitoring and adjustment. Finally, through dynamic intervention decision-making based on the construction sequence evolution curve, this invention can optimize the node installation sequence, worker scheduling, and sling force distribution in real time during construction, effectively reducing the risk of misalignment transmission, improving construction efficiency, accuracy, and safety, and providing data support and intelligent management methods for the construction of complex high-altitude structures. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the method of the present invention. Detailed Implementation

[0016] 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.

[0017] Example 1: Please refer to Figure 1 As shown in the figure, the high-altitude bulk construction method for steel space frame based on triangular rods according to an embodiment of the present invention includes the following steps: S1: In the high-altitude operation area with multi-point lifting cooperation, the triangular rod components to be assembled and the corresponding node ball components are lifted and positioned synchronously. The key spatial parameters reflecting the stability of the components are extracted with the help of the attitude sensing measurement component. The measurement results are dynamically corrected by the sling force adjustment mechanism and the high-altitude displacement compensation unit to form a continuous and traceable assembly attitude change sequence.

[0018] The process in S1 of extracting key spatial parameters reflecting the stability of the component using an attitude sensing measurement component is as follows: Three-dimensional attitude sensors and angular velocity sensors are deployed around the lifting points and nodal spheres to collect the pitch angle, yaw angle, and axial tilt angle of the triangular rod component in real time under high-altitude conditions. The parameters are then time-synchronized and filtered to serve as the spatial parameters of the component. An integrated attitude and angular velocity sensing unit is deployed at both ends and the mid-span of the triangular rod. Zero-point calibration and temperature drift calibration are completed on the ground reference platform before lifting. After lifting, a unified time reference device on site synchronizes all sensing units with a sampling frequency of no less than 100 times per second. A construction reference coordinate system is established using a total station or laser theodolite to align the sensing points with the construction coordinates. The raw data is first subjected to outlier removal and low-pass filtering, and then short-time window smoothing and drift compensation to obtain stable pitch, yaw, and axial tilt angles. These parameters are then timestamped into the attitude data record to form traceable spatial parameters. By combining the wind load on the component during hoisting with the stress distribution fed back by the sling force sensors, the attitude offset is calculated and added to the spatial parameters to form a complete attitude measurement result. An anemometer is set up on the upwind side of the work area and near the component to acquire instantaneous wind field changes. Tension sensors are set up at the end of each sling to continuously collect the force at each hoisting point. The wind field information, sling force, component self-weight, and inertial factors are input into the force-attitude correction module. According to the pre-built component equivalent stiffness and connection sequence rules, the rotation increment and displacement increment of the component under the current working condition are calculated. The increment is merged with the attitude angle obtained in the first step through the state fusion method and updated in real time to the attitude result after wind disturbance and force correction. At the same time, the correction source, hoisting point force distribution, and wind field intensity are recorded. The component design benchmark values ​​are obtained based on the structural drawings and finite element simulation results during the construction design phase, including the ideal attitude angle and force balance state of the component in the theoretical installation position. Before construction, a three-dimensional configuration including the geometric dimensions of the members, the spherical coordinates of the nodes, and the preload level of the connectors is established based on the approved structural construction drawings and fabrication details. The force and deformation calculations of typical construction stages and the final placement stage are performed using finite element analysis, and the target values ​​of pitch, yaw, torsion, etc., in the theoretical installation position, as well as the force distribution of the nodes and members, are output. The analysis results are corrected by combining material inspection and factory trial assembly data to form a version-controlled benchmark database, which is then frozen after confirmation by the chief engineer or professional person in charge and used as the sole benchmark for on-site comparison. Spatial parameters are compared with component design benchmark values ​​to extract key spatial parameters reflecting the components during assembly. A real-time comparison task with a one-to-one correspondence between component numbers is established, and the attitude results obtained in the second step are compared item by item with the benchmark database in the third step. Three levels of limits are set for allowable deviation, early warning deviation, and alarm deviation, and indicators such as pitch, yaw, axial tilt, micro-sway amplitude, deviation duration, deviation change rate, stress difference at each lifting point, and docking gap between adjacent nodes are classified and judged. When any indicator reaches the early warning or alarm limit, the maximum deviation, duration, corresponding lifting point and wind field status, and installation completion of adjacent components at that moment are automatically extracted and summarized into a set of key spatial parameters, with timestamps, data sources, and processing flow numbers attached, which serve as direct inputs for subsequent assembly attitude change sequence construction and construction sequence intervention judgment.

[0019] The process of forming a continuous and traceable sequence of assembly posture changes in S1 is as follows: Based on the key spatial parameters output by the attitude sensing measurement components, a time-sequential attitude change record is constructed. Before lifting, zero-point and temperature drift calibrations are performed on all attitude and force sensing units, and a unified coordinate and time reference are established using the construction measurement benchmark. After lifting, data such as pitch, yaw, axial tilt, micro-sway amplitude, and reliability of each measuring point are collected at a frequency of no less than 100 times per second. Each record is accompanied by a timestamp, component number, node number, sensor number, calibration version number, and work condition label. Outlier removal, short-term smoothing, and missing measurement completion are performed on multi-source data, and then resampling is performed at fixed time intervals to ensure no overlap or gaps between adjacent moments. When the work condition switch between the lifting stage, positioning stage, and temporary fixing stage is detected, stage markers and operator signature information are written into the record. All records are saved to the anti-tampering log by appending and are synchronized to the on-site host and cloud backup, forming a time-sequential, complete, and traceable attitude change record. The system combines a sling force adjustment mechanism to correct instantaneous attitude deviations caused by uneven lifting points or wind interference in real time. Tension sensors and length feedback devices are installed on each sling to continuously monitor the difference between the maximum and minimum forces at each lifting point, as well as the rate and duration of force change. When the force imbalance exceeds the warning threshold, a micro-speed adjustment or micro-release command is issued to the corresponding hoisting mechanism to balance the force. When sudden changes in wind speed or direction trigger disturbance events, a sway suppression strategy is activated, including a short-term reduction in hoisting speed, a brief pause to allow for attenuation, and, if necessary, simultaneous staggered release and resuscitation of adjacent lifting points. During the correction process, each command, execution result, force recovery status, and remaining sway are recorded in the attitude log, and a correction mark is added. If multiple corrections fail to restore the system to the allowable range within a short period, an event entry requiring manual review is automatically generated and pushed to the supervisor and chief engineer's terminals. Simultaneously, this time period is designated as a key review area in the log. A high-altitude displacement compensation unit is introduced to dynamically correct abnormal fluctuation points in the recorded sequence, forming an assembly attitude change sequence. Total stations or laser rangefinder reflective targets, ultra-wideband positioning base stations, or similar high-altitude displacement measurement devices are deployed in the assembly area to establish static reference points and dynamic tracking points. The attitude records are evaluated point-by-point, and abnormal fluctuation points are determined based on amplitude thresholds, rate of change thresholds, duration thresholds, and recurrence frequency thresholds. For single-point sudden anomalies, nearest-time interpolation and multi-source cross-verification are prioritized for correction. For continuous anomalies or consistent anomalies from multiple sources, a combination of reference point re-measurement and manual verification is used, and the processing conclusions and responsible personnel information are written into the remarks field. All corrected data retains the original value, corrected value, and correction reason, and overwriting of the original record is prohibited. Finally, the assembly attitude change sequence, including the original attitude, force correction amount, displacement compensation amount, quality identifier, and event tag, is output in a timeline format.

[0020] S2: Perform state tracking of cross-node connection dependency chains on the assembly posture change sequence, and combine historical installation cycle records, adjacent node docking success rate and component vibration attenuation law to construct a multi-dimensional structural feature vector that reflects the construction sequence and spatial constraint relationship.

[0021] The process of performing cross-node connection dependency chain state tracking on the assembly attitude change sequence in S2 is as follows: The key node attitude parameters in the assembly attitude change sequence are mapped to node connection dependency chains, marking the temporal correlation between each node; the construction BIM model or design connection table is called to obtain the component number, node number, design connection relationship, and connection type (e.g., bolt, pin, weld); the attitude records of the corresponding components are extracted from the assembly attitude change sequence according to the timestamp, and associated with the unique component / node identifier; the topological adjacency relationship defined in the design connection table is used as the initial dependency chain; if there is a slight deviation between the on-site BIM coordinates and the real-time attitude measurement points, spatial proximity (e.g., the center distance between nodes is less than the preset space) is used as the initial dependency chain. The candidate set of actual connection pairs is determined by a combination of thresholds and time correlation (e.g., component arrival time difference within a short window). In cases with multiple candidates, priority is ranked by comparing arrival time, attitude angle consistency, and connector type priority to determine the primary connection relationship and record secondary candidates as alternatives. A time-series label sequence (e.g., "awaiting arrival → in-place detection → initial assembly and fixing → final locking → acceptance") is generated for each connection dependency chain, and the corresponding time points in the assembly attitude sequence are mapped to these stages. The source sensor, timestamp, and processing version number are recorded at each stage to form an auditable time-series association mark. By combining historical installation cycle records, the actual completion rate of the installation sequence between nodes is compared; the average installation time, typical step time distribution, and standard operation cycle are retrieved from the project construction monitoring database or historical project archives for similar components / node types; if the project has early completion records, the historical cycle of this project is used as a reference; for each node, the "planned completion progress ratio" and "actual completion progress ratio" are calculated according to the time sequence label, and the actual arrival time, initial installation time, and locking time are compared with the median and upper / lower quartile times of the historical cycle to obtain completion indicators (e.g., described as percentages or grades); three levels of deviation strategies are preset (normal, warning, abnormal), for example, when the actual completion time is later than the reference median by a certain multiple or exceeds the reference upper quartile, a warning is issued, and exceeding a higher multiple or exceeding the maximum allowable delay is considered abnormal; at the same time, the amplification effect of continuous delay and multi-node series delay is considered to perform a weighted evaluation of the overall completion of the cross-node chain; The success and failure rates of adjacent node docking are statistically analyzed. Combined with the vibration attenuation patterns of components during assembly, the evolution path of cross-node dependency chains is dynamically tracked, outputting a state trajectory that simultaneously reflects temporal and physical constraints. The criteria for successful docking are clearly defined in the design specifications or construction standards (e.g., gap below allowable value, attitude angle deviation below allowable angle, pre-tightening of connectors reaching nominal torque without abnormal stress concentration). These criteria are then mapped to measurable items (gap measurement, attitude deviation, bolt torque reading, stress sensor output). Failure is indicated by any key criterion exceeding the alarm limit or existing... If initial fixing is not completed due to installation difficulties, maintain a rolling window for each pair of adjacent nodes (e.g., recent docking attempts or attempts over recent hours / days) to calculate the number of successful attempts, failures, success rate, and failure cause distribution. Simultaneously, analyze typical failure modes (e.g., angle deviation, gap exceeding limits, uneven stress) to identify recurring problems. Extract vibration indicators (e.g., peak amplitude, peak-to-peak amplitude, and time required for amplitude to decay to a certain proportion) from the assembly attitude change sequence. Use the vibration decay characteristics measured on-site as a health reference; if the vibration decays rapidly and within the allowable time... Stability within the window indicates that physical constraints are met; if vibration decays slowly or continues to exceed limits, it indicates that the connection stiffness may be insufficient or the force transmission may be abnormal. Vibration judgment factors also combine force sensor readings (sling stress fluctuations), wind field records, and operational events (such as lifting / lowering / locking actions); within the system, "evolution path objects" are maintained on a chain-by-chain basis, recording the state changes of each node (awaiting connection → contact → initial determination → reinforcement → acceptance / fault pending handling), and physical constraint scores (based on a comprehensive calculation of vibration decay, force balance, gap deviation, etc.) and timing compliance scores are added to the path. Based on historical beat comparison); the path object is appended with a timestamp and human / machine action information each time its state changes; the system outputs a time-series state trajectory report by chain, including the node status, attitude deviation, docking judgment (success / failure / pending confirmation) at each moment, vibration index, force distribution snapshot, completion comparison and recommended operation (e.g., "continue fixing", "adjust sling stress", "manual review"); when a series of risks occur on the path (e.g., multiple key nodes approaching the abnormal threshold at the same time), the system generates a chain-level warning and suggests suspending subsequent downstream docking or prioritizing local reinforcement.

[0022] The process of constructing a multidimensional structural feature vector reflecting the construction sequence and spatial constraints in S2 is as follows: Map the attitudes, docking results, and component stress states of each key node in the cross-node dependency chain state trajectory to the feature space; extract the original observation items corresponding to each individual node from the cross-node dependency chain state trajectory, including but not limited to: unique node number, timestamp, attitude angle (pitch, yaw, torsion), relative position deviation, docking gap / fitting amount, whether the joint is completed and the completion time, tension values ​​at each lifting point, torque readings of connecting parts, peak vibration and decay time, time difference between adjacent nodes, on-site environmental quantities (instantaneous wind speed, wind direction, temperature), and operator / team identification; preprocess the original observation items according to a unified field template: advanced The process involves: aligning all records with a unified time reference; removing obviously invalid readings (e.g., over-range or sensor fault markers); filling short-term missing measurements with time-series interpolation or adjacent sensor regression; and adding a source identifier and sensor calibration version number to each record. The preprocessed raw observations are mapped to several atomic features (e.g., positioning delay, absolute angle deviation, force imbalance amplitude, vibration energy index, docking duration, etc.), and statistical semantics (instantaneous value, short-term mean, short-term variance, peak value, duration of continuous over-limit, etc.) are recorded for each atomic feature to form a node-level feature list, serving as the original dimension set for entering the feature space. By integrating operator cycle time, lifting point stress distribution, and component spatial constraints during construction, a multi-dimensional statistical parameter set is formed. Based on node-level features, neighborhood summary features are constructed along the construction chain: for example, statistics on the arrival rate, average installation time, cumulative delay, and failure rate of upstream and downstream nodes within a set time window; for lifting point stress, calculations of the stress difference distribution, maximum to minimum stress ratio, and stress fluctuation frequency of all lifting points at that node; for operator cycle time, summarizing the historical average working time of the same work group on similar components, the current work group's experience level, and real-time attendance; and for component spatial constraints, reading the design BIM. Alternatively, geometric constraints in the process drawings (such as minimum clearance between nodes, maximum allowable angular deviation, and connector tolerance grade) are used, and the displacement and angular deviations measured on-site are compared with these design constraints to extract the frequency of exceedances, the distribution of exceedance amplitudes, and the types of connectors affected. The above node-level and neighborhood-level statistics are aggregated into a multi-dimensional statistical parameter set by category, including: time-series characteristics (arrival rate, cycle time deviation), spatial constraint characteristics (position deviation, fit clearance exceedance rate), stress / stability characteristics (stress imbalance, vibration damping performance), quality event characteristics (historical failure rate, number of repairs), and personnel / organization characteristics (team efficiency, personnel rotation frequency). By processing the parameter set through normalization and feature reduction algorithms, a multi-dimensional structural feature vector that can reflect the rationality of the construction sequence and the degree of spatial constraint tension is generated.

[0023] S3: Generate an assembly sequence evolution curve from multi-dimensional structural feature vectors. Based on the degree of deviation between the assembly sequence evolution curve and the actual component arrival time sequence, identify the node positions with installation misalignment risks and assign priority to the assembly links corresponding to the node positions.

[0024] The process of identifying node locations at risk of installation misalignment in S3 is as follows: Based on multidimensional structural feature vectors, an assembly sequence evolution curve is generated to extract the change pattern of the construction sequence. The multidimensional structural feature vector sequence corresponding to the cross-node dependency chain is read periodically (e.g., every second or triggered by an event) from the construction database. Missing fields are first filled using nearest neighbor interpolation or historical median, and the filling method and confidence level are recorded. Numerical features are robustly normalized (e.g., based on historical median and interquartile range), and categorical features are uniformly encoded. All records retain timestamps, chain identifiers, and processing version numbers. The multidimensional vectors at each time point are sorted by time, and the multidimensional information is mapped into one or more representative time-series curves according to preset rules (e.g., "sequence consistency score curve," "spatial tension curve," "stress stability curve"). The mapping method can employ weighted linear aggregation (assigning weights to key components and summing them) or output a single-index time series through a pre-trained model (such as a supervised regression model or an unsupervised dimensionality reduction model); the mapping process specifies the source of weight values ​​or the source of model training data in the specification; the output time series curve is applied with segmented / change point detection methods (such as sliding window differencing, steady-state window comparison, or threshold-based abrupt change detection) to identify patterns such as "stable segment", "slow drift segment", "abrupt segment", "oscillation segment", and "stagnation segment"; the start and end times, duration, average value, and rate of change of each segment are recorded, and each pattern is assigned engineering semantics (e.g., "slow drift" may indicate stress accumulation, and "abrupt change" may indicate instantaneous disturbance or abnormal installation operation); The assembly sequence evolution curve is compared with the actual component arrival time sequence to calculate the offset index; the planned arrival time sequence of components / nodes is read from the construction schedule or BIM plan, prioritizing the version of the plan issued by the project; the actual arrival timestamps are read from the site system (based on the time of stages such as "first contact", "initial installation and fixing", and "final locking"), and the legality of the arrival events is verified (e.g., the same component should not have time reversal, and duplicate events should be merged); the planned arrival sequence and the actual arrival sequence are aligned in time according to the chain node order; for cases with slight sequence differences, a sliding window local comparison is used to identify local reordering; for larger nonlinear deviations, a sequence distance metric (such as a configured sorting gap metric or dynamic alignment method) can be used to measure the degree of inconsistency between the two sequences; the offset (time offset and sequence offset) is calculated by combining the two dimensions. Time offset part: Calculate the normalized average absolute difference between the actual arrival time and the planned arrival time (considering the allowable delay for each type of node), and map it to a normalized score of 0-1; Sequence offset part: Count the number of inversions or local rearrangements of node pairs and normalize them to a score of 0-1; The time offset score and the sequence offset score are combined into an overall offset index according to a preset weight (for example, the weight can be set to 0.6 for time offset and 0.4 for sequence offset by default, but can be calibrated in actual engineering). The offset level threshold is set in the system configuration (example: offset ≤ 0.2 is normal, 0.2 < offset ≤ 0.5 is a warning, and offset > 0.5 is an abnormality), and the time point that triggers the threshold and the set of nodes involved are recorded. When the offset exceeds the preset threshold, the location of the critical node with potential installation misalignment risk is located by combining abnormal fluctuations in node attitude and docking delay signals.

[0025] The process of prioritizing the assembly links corresponding to the node positions in S3 is as follows: Identified installation misalignment risk nodes and their upstream and downstream dependent chains are designated as key monitoring targets. Node IDs, spatial coordinates, risk scores, and trigger time windows are obtained from the analysis results of the identified misalignment risk nodes. Based on the local construction relationship map, the direct upstream and downstream nodes, key dependent chains, and component connection information related to each risk node are extracted. Chains containing multiple risk nodes or high-risk score nodes are prioritized and sorted according to the work sequence and actual construction progress to form a set of key monitoring chains. A node list, connection relationships, dependency directions, chain length, and preliminary risk descriptions of associated risk nodes are generated for each chain. Based on the construction safety level, component span size, and stress transmission capacity of the nodes, the assembly link is graded and assessed, and different priority labels are assigned to the assembly link according to the assessment results. Construction design documents and construction safety specifications are read to obtain the safety level corresponding to each node or component (such as the risk factor of high-altitude operations, the complexity of hoisting operations, and the requirements for emergency braking or support measures). The span length, weight, and material properties of each component in the chain are statistically analyzed, and components with larger spans or those bearing critical structural functions are marked as high-risk factors. Based on finite element analysis or historical construction data from the construction design stage, the maximum stress, critical torque, and stress distribution that the nodes can withstand are obtained. The actual load-bearing capacity estimate of the nodes is corrected by combining attitude measurement and force sensor feedback. Each node in each chain is assigned a value according to three dimensions: safety level, span size, and stress transmission capacity (using a score of 1-5 or low / medium / high grading), and combined with risk assessment. The scoring system generates a weighted comprehensive score for each chain. This comprehensive score is then mapped to a priority label. For example, a high-risk chain with a score ≥ threshold A is labeled "high priority," a medium-risk chain with a score between threshold B and A is labeled "medium priority," and a low-risk chain with a score ≤ threshold B is labeled "low priority." Thresholds A and B can be set based on on-site experience or specifications and can be dynamically adjusted. Each chain is labeled with a priority label in the construction management system or local construction relationship diagram, while also recording the contribution score, score composition, calculation time, and evaluation factors of the risk nodes within the chain. The priority chain information is output as structured data (node ​​ID, chain ID, priority label, score details) for use by the construction scheduling, real-time monitoring, and dynamic intervention decision-making modules. Simultaneously, a visual highlighting diagram is generated, marking high-priority chains with different colors in the 3D construction drawing to facilitate quick identification of key chains by on-site operators.

[0026] S4: The local grid area and stress transmission range where the installation misalignment risk node is located are divided into the intervention work area. The layout of workers, the stress state of the slings and the installation completion of adjacent components in the intervention work area are integrated to establish a local construction relationship map with real-time update capability.

[0027] The process in S4 of dividing the local grid area where the installation misalignment risk node is located and the stress transmission range into the intervention work area is as follows: Based on the spatial coordinates of the nodes at risk of installation misalignment, the local space frame area where the coordinates are located is determined. The three-dimensional spatial coordinates of the nodes, including their X, Y, and Z positions and node numbers, are obtained from the analysis of nodes identified as having installation misalignment risks. According to the space frame design drawings and construction BIM model, the space frame structure is divided into several functional blocks (such as span sections, truss units, or node clusters). Based on the node coordinates, the space frame unit or grid block in which the node belongs is determined, and the risk nodes are mapped to the corresponding local space frame area. The block number and the node's position within the block are recorded. A local space frame area is generated, containing the block information to which each risk node belongs. Based on structural mechanics analysis, the stress transmission range corresponding to the misaligned node is calculated; the finite element model or mechanical analysis report from the construction design stage is read to obtain the force path of the node, the force distribution of the connecting components, and the load-bearing capacity of the node; combined with on-site attitude measurement data, sling force sensor data, and wind load and weight load information, the stress transmission range of the misaligned node under actual construction conditions is calculated, including the influence distance along the upstream and downstream trusses and adjacent nodes; the theoretical stress transmission range is adjusted according to the node installation sequence and the status of adjacent nodes in actual construction to make the calculation results closer to the actual on-site stress; a list of stress influence areas for each risk node is generated, including the affected nodes, the degree of influence, and the direction of force propagation. The spatial extent of the local space frame area is merged with the stress-affected area to form the boundary of the work area to be intervened. The boundary of the local space frame area is superimposed with the stress transmission range of the nodes to obtain a three-dimensional spatial area containing all potentially affected components and nodes. Based on the merged spatial information, a three-dimensional boundary outline of the work area to be intervened is formed, which can be visualized and annotated in the construction BIM model. Each component and node in the work area to be intervened is assigned a unique identifier, and its belonging relationship in the local space frame area and stress transmission range is recorded, providing clear operational targets for construction sequence intervention and sling stress control.

[0028] The process of establishing a local construction relationship map with real-time updating capability in S4 is as follows: Key elements are extracted from the work area to be intervened, including the layout of workers, the stress state of slings, and the completion status of adjacent components. The distribution location, work group, and job responsibilities of on-site workers are read from the construction scheduling system. Combined with data from the construction site positioning system, the spatial location and movement trajectory of personnel in the work area to be intervened are obtained in real time. Through force sensors and tension monitoring devices deployed at lifting points and component connections, the tension, force direction, and force change trend of each sling are obtained in real time. Based on the on-site BIM model and high-altitude attitude measurement data, the installation status of each adjacent component in the work area to be intervened is identified to determine whether the component is in place, whether the installation is complete, and the docking accuracy with surrounding components. Based on the actual construction progress and work scheduling plan, establish logical connections and physical constraints between nodes; regard key components, node spheres, sling connection points, and worker positions within the work area to be intervened as graph nodes; construct directed logical edges between nodes according to the construction sequence, installation dependencies, construction task allocation, and sling force path to represent the sequence and dependency constraints between nodes; combine the spatial location of components, force distribution, and space frame structure design parameters to establish physical constraints between nodes, representing the construction sequence or synchronization rules that must be followed due to mechanical or spatial limitations; By updating the relationship map through real-time monitoring data, the constantly changing working environment and structural status during construction are reflected, forming a local construction relationship map.

[0029] S5: Based on the local construction relationship map, combined with the priority calibration results and the hoisting micro-sway characteristics presented in the assembly posture change sequence, dynamic intervention decision is made on the construction sequence evolution curve.

[0030] The process of implementing dynamic intervention decision-making on the construction sequence evolution curve in S5 is as follows: By combining the priority calibration results in the local construction relationship diagram with the hoisting micro-sway characteristics in the attitude change sequence, the rationality of the current construction sequence is comprehensively evaluated. Priority labels, dependent chain information, and node attributes of each node and its assembly link are read from the local construction relationship diagram. Data from the high-altitude attitude measurement component and sling force sensors are used to statistically analyze the amplitude, frequency, and duration of micro-sway of each node in the assembly attitude change sequence. Correlation analysis is performed between the priority calibration results and the node micro-sway characteristics to identify vibration concentration areas, abnormal sling force areas, or potential component misalignment propagation chains that may be caused by an unreasonable construction sequence. A construction sequence rationality assessment report is generated, including the risk level of each node or chain, a list of abnormal micro-sway nodes, and potential misalignment propagation paths. When potential risks of misalignment are detected in the construction sequence, the installation order is adjusted based on the risk level, and the scheduling of workers and the distribution of sling stress are optimized. Based on the rationality assessment report, high-risk nodes and their dependent chains are marked as intervention targets. According to the node risk level, construction dependencies, component stress limitations, and construction scheduling constraints, the installation order of high-risk nodes and their upstream and downstream nodes is adjusted to ensure that key nodes are installed first or simultaneously to reduce the transmission of misalignment risks. An adjusted construction sequence plan is generated, including the new installation order of each node, the estimated completion time, and the affected dependent chains, providing a basis for on-site construction and monitoring system execution.

[0031] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0032] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A high-altitude bulk construction method for a steel structure net rack based on a triangular pole, characterized in that, The method comprises the following steps: Synchronous hoisting and positioning of the triangular rod member to be assembled and the corresponding node ball in the high-altitude operation area of multi-hoisting points, extraction of key spatial parameters reflecting the stability of the member by means of the attitude sensing measurement assembly, dynamic correction of the measurement results by means of the hoisting cable force adjusting mechanism and the high-altitude displacement compensation unit, and formation of a continuous and traceable assembly attitude change sequence; State tracking of the cross-node connection dependency chain of the assembly attitude change sequence, construction of a multi-dimensional structure feature vector reflecting the construction sequence and spatial constraint relationship in combination with the historical installation rhythm record, the successful rate of adjacent node docking and the vibration attenuation law of the member; Generation of an assembly sequence evolution curve from the multi-dimensional structure feature vector, identification of the node positions with installation misplacement risks according to the deviation degree of the assembly sequence evolution curve and the actual member time sequence, and priority rating of the assembly link corresponding to the node positions; Division of the local grid area where the installation misplacement risk node positions are located and the stress conduction range into a to-be-intervened operation area, and establishment of a local construction relationship graph with real-time updating capability by fusing the operation personnel arrangement in the to-be-intervened operation area, the hoisting cable force state and the installation completion degree of the adjacent member; Dynamic intervention decision of the construction sequence evolution curve in combination with the priority rating result and the hoisting micro-oscillation characteristics presented in the assembly attitude change sequence based on the local construction relationship graph.

2. The steel structure space truss high-altitude bulk construction method based on triangular poles according to claim 1, characterized in that, The process of extracting key spatial parameters reflecting the stability of the member by means of the attitude sensing measurement assembly is as follows: Real-time collection of the pitch angle, yaw angle and axial inclination of the triangular rod member in the high-altitude environment by means of the three-dimensional attitude sensor and angular velocity sensor arranged around the hoisting point and the node ball, and parameter time synchronization and filtering processing, so that the parameters are used as the spatial parameters of the member; Calculation of the attitude deviation in combination with the wind load borne by the member during hoisting and the stress distribution feedback of the hoisting cable force sensor, and supplement of the attitude deviation to the spatial parameters, so that complete attitude measurement results are formed; The member design reference value is obtained according to the structure drawing and the finite element simulation calculation result in the construction design stage, and includes the ideal attitude angle and the force balance state of the member in the theoretical installation position; Comparison of the spatial parameters with the member design reference value, and extraction of the key spatial parameters reflecting the spatial change of the member in the assembly process.

3. The steel structure space truss high-altitude bulk construction method based on triangular poles according to claim 2, characterized in that, The process of forming a continuous and traceable assembly attitude change sequence is as follows: Construction of the attitude change record arranged in time sequence based on the key spatial parameters output by the attitude sensing measurement assembly; Real-time correction of the instantaneous attitude deviation caused by the unbalance of the hoisting point or the wind field interference in combination with the hoisting cable force adjusting mechanism; Dynamic correction of the abnormal fluctuation points in the record sequence by means of the high-altitude displacement compensation unit, so that the assembly attitude change sequence is formed.

4. The steel structure space truss high-altitude bulk construction method based on triangular poles according to claim 3, characterized in that, The process of state tracking of the cross-node connection dependency chain of the assembly attitude change sequence is as follows: Mapping of the key node attitude parameters in the assembly attitude change sequence to the node connection dependency chain, and marking of the time sequence correlation between the nodes; Comparison of the actual completion degree of the installation sequence between the nodes in combination with the historical installation rhythm record; The success rate and failure rate of the abutment of adjacent nodes are counted, the vibration attenuation law of the components in the assembly process is combined, the evolution path of the cross-node dependent chain is dynamically tracked, and the state trajectory reflecting the timing and physical constraints is output.

5. The steel structure space truss high-altitude bulk construction method based on triangular poles according to claim 4, characterized in that, The process of constructing the multi-dimensional structural feature vector reflecting the construction sequence and spatial constraint relationship is as follows: Map the key node posture, abutment result and component stress state in the cross-node dependent chain state trajectory to the feature space; Fusion the operation tempo of the workers in the construction process, the force distribution of the lifting points and the spatial position constraint parameters of the components to form a multi-dimensional statistical parameter set; Through normalization and feature dimension reduction algorithm, the multi-dimensional structural feature vector reflecting the rationality of the construction sequence and the tightness of the spatial constraint is generated.

6. The steel structure space truss high-altitude bulk construction method based on triangular poles according to claim 5, characterized in that, The process of identifying the node position with installation misplacement risk is as follows: Generate the assembly sequence evolution curve based on the multi-dimensional structural feature vector, extract the change mode of the construction sequence; Compare the assembly sequence evolution curve with the actual component time sequence, calculate the deviation index; When the deviation exceeds the preset threshold, combined with the abnormal fluctuation of the node posture and the delay signal of the abutment, the key node position with potential installation misplacement risk is located.

7. The steel structure space truss high-altitude bulk construction method based on triangular poles according to claim 6, characterized in that, The process of prioritizing the assembly link corresponding to the node position is as follows: Take the identified installation misplacement risk node and its upstream and downstream dependent chain as the key monitoring object; Combined with the construction safety level, the component span size and the node stress conduction capacity, the assembly link is graded and evaluated, and different priority labels are given to the assembly link according to the evaluation results.

8. The steel structure space truss high-altitude bulk construction method based on triangular poles according to claim 7, characterized in that, The process of dividing the local grid area where the installation misplacement risk node position is located and the stress conduction range into the intervention work area is as follows: Based on the position coordinates of the installation misplacement risk node in space, determine the local grid area where the position coordinates are located; Combined with structural mechanics analysis, calculate the stress conduction range corresponding to the misplacement node; Fuse the spatial range of the local grid area with the stress influence area to form the boundary of the intervention work area.

9. The steel structure space truss high-altitude bulk construction method based on triangular poles according to claim 8, characterized in that, The process of establishing a local construction relationship graph with real-time updating capability is as follows: Extract the key elements from the intervention work area, including the arrangement of workers, the force state of lifting ropes and the installation completion degree of adjacent components; According to the actual construction progress and work scheduling plan, establish the logical association and physical constraint relationship between nodes; Update the relationship graph through real-time monitoring data to reflect the changing work environment and structure state in the construction process, and form the local construction relationship graph.

10. The steel structure space truss high-altitude bulk construction method based on triangular poles according to claim 9, characterized in that, The process of implementing dynamic intervention decision on the construction sequence evolution curve is as follows: Combined with the priority labeling results in the local construction relationship graph and the lifting micro-swing characteristics in the posture change sequence, comprehensively evaluate the rationality of the current construction sequence; When potential misplacement conduction risk is detected in the construction sequence, adjust the installation sequence based on the risk level, and optimize the worker scheduling and lifting rope force distribution.

Images