Safety Early Warning Method and Device for Roof Tube Truss Steel Structure Hoisting under Sensor Monitoring

Abstract

The application discloses a roof pipe truss steel structure hoisting safety early warning method and device under sensing monitoring and relates to the technical field of sensing monitoring. The method comprises the following steps: acquiring Q hoisting points and K steel structure stress points of a target pipe truss steel structure; continuously monitoring the Q hoisting points and the K steel structure stress points to obtain a Q hoisting point monitoring data group sequence and a K steel structure stress point monitoring group sequence; iteratively identifying risk coefficients of the Q hoisting point monitoring data group sequence and the K steel structure stress point monitoring group sequence to determine a first-level risk monitoring object and a second-level risk monitoring object; and performing asynchronous double-channel safety monitoring on the first-level risk monitoring object and the second-level risk monitoring object to obtain first safety early warning information. The technical problem that safety hazards cannot be timely and effectively early warned in the hoisting process of a roof pipe truss steel structure in the prior art is solved, and the technical effect that hoisting operation safety early warning is realized through continuous monitoring of hoisting points and stress points by sensing monitoring is achieved.

Description

Technical Field

[0001] This invention relates to the field of sensor monitoring technology, specifically to a method and device for safety early warning of roof truss steel structure hoisting under sensor monitoring. Background Technology

[0002] With the continuous expansion of construction projects and the gradual increase in construction difficulty, especially in the hoisting of large roof truss steel structures, hoisting safety has become a critical issue that urgently needs to be addressed. Traditional hoisting safety monitoring methods rely heavily on manual inspection and experience-based judgment, often failing to detect potential safety hazards in a timely manner, particularly with significant blind spots in monitoring lifting points and stress points. With the development of sensor technology, real-time monitoring using sensor arrays has become an effective way to improve hoisting safety. However, most existing sensor monitoring systems focus on monitoring single data points, lacking comprehensive multi-point and multi-dimensional assessments of the hoisting process, resulting in insufficient risk identification and early warning capabilities. Summary of the Invention

[0003] This application provides a method and device for safety early warning during the hoisting of roof truss steel structures under sensor monitoring, which solves the technical problem that safety hazards cannot be warned in a timely and effective manner during the hoisting of roof truss steel structures in the prior art.

[0004] The first aspect of this application provides a method for safety early warning during the hoisting of roof truss steel structures under sensor monitoring, the method comprising:

[0005] Q lifting points and K stress points of the target tubular truss steel structure are obtained, where Q and K are positive integers. A sensor monitoring array is used to continuously monitor the Q lifting points and K stress points, respectively, to obtain a sequence of monitoring data sets for the Q lifting points and the K stress points. The sequence of monitoring data sets for the Q lifting points and the K stress points is traversed to identify risk coefficients, determining primary and secondary risk monitoring targets. Asynchronous dual-channel safety monitoring is performed on the primary and secondary risk monitoring targets to obtain first-level safety warning information, where the asynchronous dual-channel includes a first monitoring sub-channel and a second monitoring sub-channel.

[0006] A second aspect of this application provides a safety early warning device for the hoisting of roof truss steel structures under sensor monitoring, the device comprising:

[0007] The key point acquisition module is used to acquire Q lifting points and K stress points of the target tubular truss steel structure, where Q and K are positive integers; the data acquisition module is used to continuously monitor the Q lifting points and K stress points using a sensor monitoring array to obtain a sequence of monitoring data sets for the Q lifting points and the K stress points; the risk identification module is used to traverse the sequence of monitoring data sets for the Q lifting points and the K stress points to identify risk coefficients and determine the primary and secondary risk monitoring objects; the safety monitoring module is used to perform asynchronous dual-channel safety monitoring on the primary and secondary risk monitoring objects to obtain first safety warning information, wherein the asynchronous dual channels include a first monitoring sub-channel and a second monitoring sub-channel.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] First, Q lifting points and K stress points of the target tubular truss steel structure are acquired, where Q and K are positive integers. Next, a sensor monitoring array is used to continuously monitor the Q lifting points and K stress points, obtaining monitoring data sequences for the Q lifting points and K stress points respectively. Then, the risk coefficients are identified by traversing the monitoring data sequences for both lifting points and stress points, determining the primary and secondary risk monitoring targets. Finally, asynchronous dual-channel safety monitoring is performed on the primary and secondary risk monitoring targets to obtain the first safety warning information. The asynchronous dual-channel includes a first monitoring sub-channel and a second monitoring sub-channel. This solves the technical problem in existing technologies where safety hazards during the hoisting of roof tubular truss steel structures cannot be warned in a timely and effective manner, achieving the technical effect of continuous monitoring of lifting points and stress points through sensor monitoring to realize safety warnings for hoisting operations. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A schematic diagram of the safety early warning method for hoisting roof truss steel structure under sensor monitoring provided in the embodiments of this application;

[0012] Figure 2 A schematic diagram of the safety early warning device for the hoisting of roof truss steel structure under sensor monitoring provided in this application embodiment.

[0013] Explanation of reference numerals in the attached diagram: Key point acquisition module 11, data acquisition module 12, risk identification module 13, and safety monitoring module 14. Detailed Implementation

[0014] This application provides a method and device for safety early warning during the hoisting of roof truss steel structures under sensor monitoring, which solves the technical problem that safety hazards cannot be warned in a timely and effective manner during the hoisting of roof truss steel structures in the prior art.

[0015] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0016] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0017] Example 1, as Figure 1 As shown, this application provides a method for safety early warning of roof truss steel structure hoisting under sensor monitoring, wherein the method includes:

[0018] Obtain the Q lifting points and K stress points of the target tubular truss steel structure, where Q and K are positive integers.

[0019] Based on the hoisting plan for the target tubular truss steel structure, identify Q hoisting points and K stress points of the target roof tubular truss steel structure, where Q and K are positive integers. A hoisting point is the connection point between the hoisting equipment (such as a crane or gantry crane) and the steel structure during hoisting operations; a stress point is a critical part of the steel structure that bears external loads or forces.

[0020] The sensor monitoring array is used to continuously monitor the Q lifting points and K steel structure stress points respectively, and obtain the monitoring data sequence of the Q lifting points and the monitoring data sequence of the K steel structure stress points.

[0021] Furthermore, the sensor monitoring array includes strain sensors, displacement sensors, load sensors, and accelerometers.

[0022] A sensor monitoring array is used to continuously monitor Q lifting points and K steel structure stress points, thereby acquiring relevant data from these points in real time for dynamic assessment of the structural condition during hoisting. Specifically, appropriate types of sensors are installed at each lifting point and stress point. Lifting points typically require strain sensors, displacement sensors, and load sensors to monitor mechanical changes during hoisting, such as displacement, stress, and load variations. Steel structure stress points require strain sensors, accelerometers, and other sensors to detect the stress state and deformation of the structure in real time. Once the sensor array is installed, it continuously monitors the Q lifting points and K steel structure stress points in real time. The sensors at each lifting point and stress point periodically collect data and send it to the monitoring system, which organizes this data into data sets. For each lifting point, Q lifting point monitoring data sets are obtained, and for each stress point, K steel structure stress point monitoring data sets are obtained. These monitoring data sets contain real-time changes in lifting points and stress points during the hoisting process, such as stress, displacement, vibration, and load. This data will provide a foundation for subsequent risk assessment, anomaly monitoring, and early warning. Through this continuous monitoring data, the dynamic changes of the structure during hoisting can be accurately grasped, thereby promptly identifying potential safety risks and hidden dangers and ensuring the smooth progress of hoisting operations.

[0023] Risk coefficients are identified by traversing the Q sets of monitoring data for lifting points and the K sets of monitoring data for stress points of steel structures, thereby determining the primary and secondary risk monitoring targets.

[0024] By traversing the monitoring data sequences of Q lifting points and K steel structure stress points, the system analyzes and processes the monitoring data for each lifting point and stress point. Based on this data, a risk coefficient is calculated for each lifting point and stress point. The calculation method for the risk coefficient includes a comprehensive assessment of multiple factors, such as the volatility of the monitoring data, the magnitude of change, and the degree of deviation from the safety threshold. Specifically, for each lifting point and stress point, the volatility factor and correlation factor of its data sequence can be calculated, and the risk coefficient of that lifting point or stress point is obtained through weighted calculation. The risk coefficient reflects the degree of potential safety hazards at that point during the lifting process.

[0025] After calculating the risk coefficients, the system will classify and identify the Q lifting points and K steel structure stress points according to the preset risk coefficient thresholds. Based on the risk coefficient, the monitored objects are divided into Level 1 and Level 2 risk monitoring objects. Level 1 risk monitoring objects refer to those points with high risk coefficients, indicating that there are significant safety hazards during the lifting process, which may lead to structural instability, overloading of lifting equipment, and other problems. Level 2 risk monitoring objects refer to those points with lower risk coefficients, but still have certain potential hazards. Usually, the risk coefficient of these points does not reach the Level 1 risk threshold, but they still need to be monitored.

[0026] Furthermore, risk coefficients are identified by traversing the Q sets of monitoring data for lifting points and the K sets of monitoring data for steel structure stress points, thus determining the primary and secondary risk monitoring targets, including:

[0027] Fluctuation factors are identified for the Q sets of monitoring data at lifting points and the K sets of monitoring data at steel structure stress points, respectively, to obtain Q fluctuation factors for lifting point monitoring and K fluctuation factors for steel structure stress points monitoring. Iterative correlation identification of data features is performed on the Q sets of monitoring data at lifting points and the K sets of monitoring data at steel structure stress points, respectively, to obtain Q correlation factors for lifting point monitoring and K correlation factors for steel structure stress points monitoring. Weighted calculations are then performed on the Q fluctuation factors and Q correlation factors for lifting points, as well as the K fluctuation factors and K correlation factors for steel structure stress points monitoring, respectively, to obtain Q risk coefficients for lifting points and K risk coefficients for steel structure stress points. The Q risk coefficients for lifting points and K risk coefficients for steel structure stress points are then identified according to preset risk coefficient thresholds to obtain primary and secondary risk monitoring objects.

[0028] Preferably, firstly, fluctuation analysis is performed on the monitoring data sequences of Q lifting points to identify the fluctuation of data at each lifting point and obtain Q lifting point monitoring fluctuation factors. Similarly, fluctuation analysis is performed on the monitoring data sequences of K steel structure stress points to identify the fluctuation of data at each stress point and obtain K steel structure stress point monitoring fluctuation factors. The fluctuation factors reflect the amplitude and instability of data changes at the lifting points and stress points during the lifting process. Next, iterative correlation identification of data features is further performed. By progressively traversing each lifting point monitoring data sequence and stress point monitoring data sequence, key features in the data are extracted, and through iterative correlation analysis, Q lifting point monitoring correlation factors and K steel structure stress point monitoring correlation factors are identified. The lifting point monitoring correlation factors reveal the interaction relationship between lifting points, and the steel structure stress point monitoring correlation factors reveal the interaction relationship between steel structure stress points. Then, combining the fluctuation factors and correlation factors of Q suspension points, a weighted calculation is performed according to a preset weight allocation to obtain the risk coefficients of Q suspension points. The weight allocation can be set according to factors such as the importance of the suspension points in the structure and historical data performance. Similarly, combining the fluctuation factors and correlation factors of K steel structure stress points, a weighted calculation is performed to obtain the risk coefficients of K steel structure stress points. Finally, the risk coefficients of the Q suspension points and K steel structure stress points are identified and classified according to preset risk coefficient thresholds. If the risk coefficient of a certain suspension point or stress point exceeds the preset first-level risk threshold, the point is marked as a first-level risk monitoring object, indicating that the point has a high safety hazard and needs to be monitored and dealt with first. If the risk coefficient of a certain suspension point or stress point is between the first-level and second-level risks, the point is marked as a second-level risk monitoring object, indicating that the point still has some safety hazards, but the danger is lower than that of the first-level risk object, and it still needs to be monitored, but the priority is relatively low.

[0029] Furthermore, fluctuation factors are identified for the Q sets of monitoring data from the lifting points and the K sets of monitoring data from the steel structure stress points, respectively, to obtain the fluctuation factors for the Q lifting point monitoring and the K sets of monitoring data from the steel structure stress points, including:

[0030] The fluctuation variance is calculated by traversing the Q sets of monitoring data for lifting points and the K sets of monitoring data for steel structure stress points to obtain Q first-point monitoring fluctuation factors and K first-steel structure stress point monitoring fluctuation factors. The maximum amplitude comparison analysis is then performed on the Q sets of monitoring data for lifting points and the K sets of monitoring data for steel structure stress points to obtain Q second-point monitoring fluctuation factors and K second-steel structure stress point monitoring fluctuation factors. Weighted calculations are then performed on the Q first-point monitoring fluctuation factors, the Q second-point monitoring fluctuation factors, and the K first-steel structure stress point monitoring fluctuation factors and the K second-steel structure stress point monitoring fluctuation factors to obtain Q lifting point monitoring fluctuation factors and K steel structure stress point monitoring fluctuation factors.

[0031] Specifically, the monitoring data sequences of Q lifting points and K steel structure stress points are traversed. For each lifting point and steel structure stress point, the fluctuation variance is calculated. The fluctuation variance is an indicator that measures the amplitude of data fluctuation and can reflect the change range and stability of the monitoring data. For each lifting point and each stress point, the variance of its monitoring data sequence is calculated to obtain the fluctuation variance of each lifting point and stress point, thereby obtaining the Q first lifting point sub-monitoring fluctuation factors and the K first steel structure stress point monitoring fluctuation factors. Next, a maximum amplitude comparison analysis is performed on the monitoring data sequences of the Q lifting points and K steel structure stress points. Specifically, for each lifting point and stress point, its maximum and minimum values ​​are identified, and the difference between them is calculated to obtain the maximum amplitude at that point. The sum of the maximum amplitudes of all Q lifting points and K steel structure stress points is calculated. For each lifting point and stress point, the ratio of its maximum amplitude to the sum of all maximum amplitudes is calculated, thus obtaining the Q second-level lifting point sub-monitoring fluctuation factors and the K second-level steel structure stress point sub-monitoring fluctuation factors. Then, the first and second sub-monitoring fluctuation factors of the Q lifting points and the K steel structure stress points are weighted and calculated. Different weights are assigned according to the importance of fluctuation variance and maximum amplitude in the evaluation, thereby calculating the Q lifting point monitoring fluctuation factors and the K steel structure stress point monitoring fluctuation factors.

[0032] Furthermore, by traversing the Q sets of monitoring data for lifting points and the K sets of monitoring data for steel structure stress points, iterative association identification of data features is performed to obtain the Q sets of correlation factors for lifting point monitoring and the K sets of correlation factors for steel structure stress point monitoring, including:

[0033] Extract the first lifting point monitoring data group sequence from the Q lifting point monitoring data group sequence; traverse the first lifting point monitoring data group sequence to extract data features and obtain the first lifting point monitoring feature sequence; perform iterative association identification of data features on two adjacent first lifting point monitoring features in the first lifting point monitoring feature sequence from front to back to obtain the first lifting point monitoring feature association factor sequence; calculate the mean of the first lifting point monitoring feature association factor sequence to obtain the first lifting point monitoring association factor; perform iterative association identification of data features on the Q lifting point monitoring data group sequence and the K steel structure stress point monitoring group sequence respectively to obtain the Q lifting point monitoring association factor and the K steel structure stress point monitoring association factor.

[0034] First, from the Q sets of lifting point monitoring data sequences, the monitoring data of one lifting point is selected as the first lifting point monitoring data sequence. Each lifting point monitoring data sequence contains monitoring data from multiple moments during the lifting process. Next, the first lifting point monitoring data sequence is traversed to extract data features, i.e., representative features such as data volatility, trend changes, and periodicity are extracted. These features reflect the specific performance of the lifting point in terms of stress and deformation during the lifting process. The extracted feature data constitutes the first lifting point monitoring feature sequence. Then, iterative association identification of data features is performed on the first lifting point monitoring feature sequence. That is, in a sequential order from beginning to end, the association identification of adjacent first lifting point monitoring features is performed step by step to obtain the first lifting point monitoring feature association factor sequence. The mean of the first lifting point monitoring feature association factor sequence is calculated to obtain the first lifting point monitoring association factor, which reflects the average degree of association between adjacent features of that lifting point during the monitoring period. Repeat the above steps for the Q sets of monitoring data for lifting points, and calculate the monitoring correlation factor for each lifting point to obtain the Q sets of monitoring correlation factors for lifting points. The data processing process for the K sets of monitoring data for steel structure stress points is similar to that for the lifting point monitoring data sets. Extract the monitoring data set for each stress point, perform feature extraction, iterative correlation identification, and mean calculation to obtain the K sets of monitoring correlation factors for steel structure stress points.

[0035] Furthermore, the monitoring feature sequence of the first hoisting point is processed in a sequential order from front to back, and iterative association identification of data features is performed on adjacent monitoring features of the first hoisting point to obtain a sequence of association factors for the monitoring features of the first hoisting point, including:

[0036] Extract any two adjacent first hanging point monitoring features from the first hanging point monitoring feature sequence; perform feature similarity identification on the two first hanging point monitoring features to obtain a first hanging point monitoring feature similarity set; normalize the first hanging point monitoring feature similarity set and fill it into an initially empty matrix to obtain a first correlation matrix; perform correlation factor identification based on the first correlation matrix and the two first hanging point monitoring features to obtain first hanging point monitoring feature correlation factors; perform iterative correlation identification of data features of two adjacent first hanging point monitoring features on the first hanging point monitoring feature sequence to obtain a first hanging point monitoring feature correlation factor sequence.

[0037] From the first hoisting point monitoring feature sequence, features of any two adjacent first hoisting points are extracted sequentially from front to back. These two adjacent features represent the performance of the hoisting point at different time points. By comparing their similarity, it can be determined whether these two features are correlated. Feature similarity identification is performed on the extracted adjacent first hoisting point monitoring features to measure the degree of similarity between adjacent features. Euclidean distance, cosine similarity, or Pearson correlation coefficient can be calculated between these two features to quantify their similarity. Based on these similarity indices, a first hoisting point monitoring feature similarity set can be obtained, which contains the similarity values ​​between all adjacent feature pairs.

[0038] The Softmax formula is used to normalize the similarity set of the monitoring features of the first hanging point. Softmax transforms a set of values ​​into a probability distribution, ensuring that each value's output falls between 0 and 1, and that the sum of all values ​​is 1. The Softmax formula is as follows: ;in, This represents the similarity value of the i-th monitored feature. Yes Perform exponential calculations to increase the weight of larger values. This is the exponential sum of the similarity values ​​of all monitored features, used to normalize the similarity values ​​of all features. For each pair of adjacent monitored features in the first set of monitored feature similarities, the normalized similarity value is calculated using the Softmax formula. Specifically, for each item in the set, the Softmax formula calculates the relative magnitude between that item and other items in the set, ensuring the output meets the normalization requirements. After calculating the Softmax-normalized values, these normalized values ​​are filled into an initially empty matrix, forming the first correlation matrix. Each element of this matrix represents the normalized similarity value between different monitored features. After the similarity values ​​normalized by Softmax are filled into the matrix, each element in the matrix represents the degree of correlation between two monitored features.

[0039] Based on the obtained first correlation matrix and the extracted monitoring features of the two first hanging points, correlation factors are identified, that is, the correlation between the monitoring features is quantified by analyzing the relationship between them. For example, correlation factors can be identified by calculating the weighted average of the correlation values ​​in the matrix or by using other correlation assessment methods. Finally, the correlation factor of the first hanging point monitoring features is obtained, which quantifies the correlation strength between the two monitoring features.

[0040] The aforementioned iterative association identification of data features is performed on all adjacent monitoring features in the first lifting point monitoring feature sequence. After this series of steps, a first lifting point monitoring feature association factor sequence is obtained. This sequence reflects the feature correlation between the lifting points at various moments during the lifting process, thereby revealing the changing patterns of the lifting points over time and potential risk associations.

[0041] Furthermore, based on the first correlation matrix and the two first hanging point monitoring features, correlation factor identification is performed to obtain the first hanging point monitoring feature correlation factor, including:

[0042] The first correlation matrix is ​​convolved with the two first hanging point monitoring features to obtain two first hanging point monitoring enhancement features; the similarity between the two first hanging point monitoring enhancement features and the corresponding first hanging point monitoring features is calculated again, and the mean of the calculation results is obtained to obtain the first hanging point monitoring feature correlation factor.

[0043] First, a convolution operation is performed between the first correlation matrix and the monitoring features of the two first hanging points. The convolution operation is a process of sliding the correlation matrix as the convolution kernel across the monitoring feature data. In this process, each element in the first correlation matrix represents the similarity between features, while the monitoring features include data such as the force and displacement of the hanging points at different time points. The convolution operation effectively captures the local correlation between the monitoring features and the correlation matrix, thereby generating a higher-dimensional feature representation, namely the enhanced features of the two first hanging point monitoring features. These enhanced features integrate information from the original features and the correlation matrix during the convolution process, providing more expressive features and helping to identify more subtle patterns of change between the hanging point data.

[0044] Next, the similarity between the two convolutional enhanced features of the first lifting point monitoring and their corresponding first lifting point monitoring features is calculated. Similarity calculation can be based on various methods, such as Euclidean distance and cosine similarity, which reflect the numerical similarity between the enhanced features and the original features. Then, the mean of the calculated similarity values ​​is taken to obtain a value representing the overall relationship strength, i.e., the first lifting point monitoring feature correlation factor. The first lifting point monitoring feature correlation factor reflects the overall correlation strength between the two sets of monitoring features. A high correlation factor indicates that the changing trends of the two sets of monitoring features are similar, and there may be a strong common change pattern, which has an important impact on the safety of lifting operations; a low correlation factor indicates that the changes between these features are relatively independent, and there may be a low correlation.

[0045] Asynchronous dual-channel security monitoring is performed on the primary and secondary risk monitoring targets to obtain the first security early warning information. The asynchronous dual channels include a first monitoring sub-channel and a second monitoring sub-channel.

[0046] For identified Level 1 and Level 2 risk monitoring targets, real-time monitoring is conducted through asynchronous dual-channel safety monitoring to obtain initial safety warning information. Asynchronous dual-channel safety monitoring refers to the system dividing the monitoring task into two channels—a first monitoring sub-channel and a second monitoring sub-channel—and employing different monitoring strategies for monitoring targets with different risk levels. Specifically, Level 1 risk monitoring targets are assigned to the first monitoring sub-channel, which monitors at a higher frequency to obtain real-time monitoring data for these high-risk points and respond promptly to any anomalies. Level 2 risk monitoring targets are assigned to the second monitoring sub-channel, which monitors at a relatively lower frequency, but still ensures effective monitoring of potential risks. Specifically, different monitoring frequencies are configured for Level 1 and Level 2 risk monitoring targets based on historical lifting data and risk assessment results. Through asynchronous methods, the system can intensively monitor Level 1 risk targets in the first monitoring sub-channel, while monitoring Level 2 risk targets at a lower frequency in the second monitoring sub-channel. This effectively allocates resources and ensures that high-risk points are prioritized without increasing the computational burden.

[0047] In each monitoring sub-channel, the collected monitoring data is analyzed in real time to detect any anomalies or potential security risks. If an anomaly is detected in the first or second monitoring sub-channel, the system should immediately generate a first security warning message, which may include the specific type, location, severity of the anomaly, and recommended countermeasures.

[0048] Furthermore, asynchronous dual-channel security monitoring is performed on the primary and secondary risk monitoring targets to obtain first security early warning information, including:

[0049] Obtain a set of historical hoisting logs; traverse the set of historical hoisting logs to perform risk classification, obtaining a first-level historical hoisting log set and a second-level historical hoisting log set; traverse the first-level and second-level historical hoisting log sets to identify abnormal time windows, obtaining a first-level and second-level abnormal time window set; based on the size of the first-level and second-level abnormal time window sets, configure the monitoring frequency to obtain a first-level and second-level monitoring frequency; construct the first and second monitoring sub-channels of the asynchronous dual-channel based on the first-level and second-level monitoring frequencies.

[0050] Specifically, all relevant historical lifting logs are extracted or collected from the database. These logs should include the time, location, equipment information, operation process, monitoring data, and any event records related to safety or failure. The historical lifting log set is then traversed, and based on the risk factors recorded in the logs (such as lifting weight, height, environmental conditions, equipment condition, etc.) and event consequences (such as whether an accident, equipment damage, or personal injury occurred), the logs are divided into a primary historical lifting log set (high risk) and a secondary historical lifting log set (low risk). The primary and secondary historical lifting log sets are then traversed, and the time-series data in each log is analyzed to identify abnormal time windows in the lifting operation. An abnormal time window can be defined as the period during which monitoring data exceeds the normal range, the operation process deviates, or a safety event occurs. Based on the analysis results, the primary abnormal time window set (high-risk time window) and the secondary abnormal time window set are obtained respectively. The monitoring frequency is configured based on the size of the primary and secondary abnormal time window sets (i.e., the number and duration of abnormal time windows) and the proportion of these abnormal time windows to the overall hoisting operation time. Generally, the more abnormal time windows there are, the longer their duration is, or the higher their proportion to the overall operation time, the higher the corresponding monitoring frequency should be. The monitoring frequency can be expressed in time intervals, such as monitoring once per minute or hour, or it can be based on event triggers, such as monitoring at the start and end of each hoisting operation or when a specific event occurs. According to the configured primary and secondary monitoring frequencies, an asynchronous dual-channel monitoring system is constructed with a first monitoring sub-channel and a second monitoring sub-channel. The first monitoring sub-channel is used to monitor the primary risk monitoring objects at the primary monitoring frequency, and the second monitoring sub-channel is used to monitor the secondary risk monitoring objects at the secondary monitoring frequency.

[0051] Furthermore, a pre-set early warning feedback window is provided, and anomaly investigation is performed within the early warning feedback window to obtain the first early warning feedback information.

[0052] After constructing an effective asynchronous dual-channel security monitoring system and monitoring the primary and secondary risk monitoring objects to obtain the first security warning information, in order to ensure the effectiveness and timeliness of the warning information, it is necessary to preset a warning feedback window and conduct anomaly investigation within this window to obtain the first warning feedback information.

[0053] The early warning feedback window is a predefined time period, which is usually adjusted based on the characteristics of the lifting operation and historical data. In lifting operations, the purpose of setting up the early warning feedback window is to centrally assess the monitored risks, ensuring that anomalies can be detected and addressed promptly. The length of the early warning feedback window and the triggering conditions are flexibly adjusted based on factors such as the actual working environment, risk assessment of the lifting points and stress points.

[0054] When the system identifies potential safety risks during monitoring, it inputs the monitoring data into a preset feedback window for further investigation. During this process, the system performs detailed analysis of the data in the warning feedback window, checking for abnormal fluctuations or non-compliance with safety regulations. These anomalies may include overloading of lifting points, excessive deformation of stress points, and displacement exceeding safe limits. Through continuous monitoring and comparison of the data, the system can identify potential risks and abnormal events. After completing the anomaly investigation, the system generates a first warning feedback message based on the analysis results. This message provides a series of safety prompts and warning measures according to the identified anomalies. For example, if the monitoring data shows that the load on the lifting point exceeds a preset safety threshold, the system may issue an alarm, requiring the suspension of lifting operations and inspection of the lifting point.

[0055] In summary, the embodiments of this application have at least the following technical effects:

[0056] First, Q lifting points and K stress points of the target tubular truss steel structure are acquired, where Q and K are positive integers. Next, a sensor monitoring array is used to continuously monitor the Q lifting points and K stress points, obtaining monitoring data sequences for the Q lifting points and K stress points respectively. Then, the risk coefficients are identified by traversing the monitoring data sequences for both lifting points and stress points, determining the primary and secondary risk monitoring targets. Finally, asynchronous dual-channel safety monitoring is performed on the primary and secondary risk monitoring targets to obtain the first safety warning information. The asynchronous dual-channel includes a first monitoring sub-channel and a second monitoring sub-channel. This solves the technical problem in existing technologies where safety hazards during the hoisting of roof tubular truss steel structures cannot be warned in a timely and effective manner, achieving the technical effect of continuous monitoring of lifting points and stress points through sensor monitoring to realize safety warnings for hoisting operations.

[0057] Example 2, based on the same inventive concept as the roof truss steel structure hoisting safety early warning method under sensor monitoring in the foregoing examples, such as... Figure 2 As shown, this application provides a safety early warning device for the hoisting of roof truss steel structures under sensor monitoring, wherein the device includes:

[0058] The key point acquisition module 11 is used to acquire Q lifting points and K stress points of the target tubular truss steel structure, where Q and K are positive integers; the data acquisition module 12 is used to continuously monitor the Q lifting points and K stress points of the steel structure using a sensor monitoring array to obtain a sequence of monitoring data sets for the Q lifting points and the K stress points of the steel structure; the risk identification module 13 is used to traverse the sequence of monitoring data sets for the Q lifting points and the K stress points of the steel structure to identify risk coefficients and determine the primary risk monitoring objects and the secondary risk monitoring objects; the safety monitoring module 14 is used to perform asynchronous dual-channel safety monitoring on the primary risk monitoring objects and the secondary risk monitoring objects to obtain the first safety warning information, wherein the asynchronous dual channels include a first monitoring sub-channel and a second monitoring sub-channel.

[0059] Furthermore, the risk identification module 13 is used to perform the following method:

[0060] Fluctuation factors are identified for the Q sets of monitoring data at lifting points and the K sets of monitoring data at steel structure stress points, respectively, to obtain Q fluctuation factors for lifting point monitoring and K fluctuation factors for steel structure stress points monitoring. Iterative correlation identification of data features is performed on the Q sets of monitoring data at lifting points and the K sets of monitoring data at steel structure stress points, respectively, to obtain Q correlation factors for lifting point monitoring and K correlation factors for steel structure stress points monitoring. Weighted calculations are then performed on the Q fluctuation factors and Q correlation factors for lifting points, as well as the K fluctuation factors and K correlation factors for steel structure stress points monitoring, respectively, to obtain Q risk coefficients for lifting points and K risk coefficients for steel structure stress points. The Q risk coefficients for lifting points and K risk coefficients for steel structure stress points are then identified according to preset risk coefficient thresholds to obtain primary and secondary risk monitoring objects.

[0061] Furthermore, the risk identification module 13 is used to perform the following method:

[0062] The fluctuation variance is calculated by traversing the Q sets of monitoring data for lifting points and the K sets of monitoring data for steel structure stress points to obtain Q first-point monitoring fluctuation factors and K first-steel structure stress point monitoring fluctuation factors. The maximum amplitude comparison analysis is then performed on the Q sets of monitoring data for lifting points and the K sets of monitoring data for steel structure stress points to obtain Q second-point monitoring fluctuation factors and K second-steel structure stress point monitoring fluctuation factors. Weighted calculations are then performed on the Q first-point monitoring fluctuation factors, the Q second-point monitoring fluctuation factors, and the K first-steel structure stress point monitoring fluctuation factors and the K second-steel structure stress point monitoring fluctuation factors to obtain Q lifting point monitoring fluctuation factors and K steel structure stress point monitoring fluctuation factors.

[0063] Furthermore, the risk identification module 13 is used to perform the following method:

[0064] Extract the first lifting point monitoring data group sequence from the Q lifting point monitoring data group sequence; traverse the first lifting point monitoring data group sequence to extract data features and obtain the first lifting point monitoring feature sequence; perform iterative association identification of data features on two adjacent first lifting point monitoring features in the first lifting point monitoring feature sequence from front to back to obtain the first lifting point monitoring feature association factor sequence; calculate the mean of the first lifting point monitoring feature association factor sequence to obtain the first lifting point monitoring association factor; perform iterative association identification of data features on the Q lifting point monitoring data group sequence and the K steel structure stress point monitoring group sequence respectively to obtain the Q lifting point monitoring association factor and the K steel structure stress point monitoring association factor.

[0065] Furthermore, the risk identification module 13 is used to perform the following method:

[0066] Extract any two adjacent first hanging point monitoring features from the first hanging point monitoring feature sequence; perform feature similarity identification on the two first hanging point monitoring features to obtain a first hanging point monitoring feature similarity set; normalize the first hanging point monitoring feature similarity set and fill it into an initially empty matrix to obtain a first correlation matrix; perform correlation factor identification based on the first correlation matrix and the two first hanging point monitoring features to obtain first hanging point monitoring feature correlation factors; perform iterative correlation identification of data features of two adjacent first hanging point monitoring features on the first hanging point monitoring feature sequence to obtain a first hanging point monitoring feature correlation factor sequence.

[0067] Furthermore, the risk identification module 13 is used to perform the following method:

[0068] The first correlation matrix is ​​convolved with the two first hanging point monitoring features to obtain two first hanging point monitoring enhancement features; the similarity between the two first hanging point monitoring enhancement features and the corresponding first hanging point monitoring features is calculated again, and the mean of the calculation results is obtained to obtain the first hanging point monitoring feature correlation factor.

[0069] Furthermore, the security monitoring module 14 is used to perform the following methods:

[0070] Obtain a set of historical hoisting logs; traverse the set of historical hoisting logs to perform risk classification, obtaining a first-level historical hoisting log set and a second-level historical hoisting log set; traverse the first-level and second-level historical hoisting log sets to identify abnormal time windows, obtaining a first-level and second-level abnormal time window set; based on the size of the first-level and second-level abnormal time window sets, configure the monitoring frequency to obtain a first-level and second-level monitoring frequency; construct the first and second monitoring sub-channels of the asynchronous dual-channel based on the first-level and second-level monitoring frequencies.

[0071] Furthermore, the security monitoring module 14 is used to perform the following methods:

[0072] A pre-defined early warning feedback window is set up. Anomalies are investigated within the early warning feedback window to obtain the first early warning feedback information.

[0073] Furthermore, the data acquisition module 12 is used to perform the following methods:

[0074] The sensor monitoring array includes strain sensors, displacement sensors, load sensors, and accelerometers.

[0075] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0076] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0077] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for safety pre-warning of hoisting of roof pipe truss steel structure under sensing monitoring, characterized in that, The method comprises: obtaining Q lifting points and K steel structure stress points of a target tube truss steel structure, wherein Q and K are positive integers; continuously monitoring the Q lifting points and K steel structure stress points by using a sensor monitoring array to obtain a sequence of Q lifting point monitoring data groups and a sequence of K steel structure stress point monitoring groups; iterating through the sequence of Q lifting point monitoring data groups and the sequence of K steel structure stress point monitoring groups to identify risk coefficients, and determining a first risk monitoring object and a second risk monitoring object; performing asynchronous double-channel safety monitoring on the first risk monitoring object and the second risk monitoring object to obtain first safety warning information, wherein the asynchronous double-channel comprises a first monitoring sub-channel and a second monitoring sub-channel; iterating through the sequence of Q lifting point monitoring data groups and the sequence of K steel structure stress point monitoring groups to identify risk coefficients, and determining a first risk monitoring object and a second risk monitoring object, comprising: identifying fluctuation factors of the sequence of Q lifting point monitoring data groups and the sequence of K steel structure stress point monitoring groups respectively to obtain Q lifting point monitoring fluctuation factors and K steel structure stress point monitoring fluctuation factors; iterating through the sequence of Q lifting point monitoring data groups and the sequence of K steel structure stress point monitoring groups to identify data feature iterative correlation to obtain Q lifting point monitoring correlation factors and K steel structure stress point monitoring correlation factors; performing weighted calculation on the Q lifting point monitoring fluctuation factors and the Q lifting point monitoring correlation factors, and the K steel structure stress point monitoring fluctuation factors and the K steel structure stress point monitoring correlation factors respectively to obtain Q lifting point risk coefficients and K steel structure stress point risk coefficients; identifying the Q lifting point risk coefficients and the K steel structure stress point risk coefficients according to a preset risk coefficient threshold to obtain a first risk monitoring object and a second risk monitoring object; identifying fluctuation factors of the sequence of Q lifting point monitoring data groups and the sequence of K steel structure stress point monitoring groups to obtain Q lifting point monitoring fluctuation factors and K steel structure stress point monitoring fluctuation factors, comprising: iterating through the sequence of Q lifting point monitoring data groups and the sequence of K steel structure stress point monitoring groups to calculate fluctuation variances to obtain Q first lifting point sub-monitoring fluctuation factors and K first steel structure stress point sub-monitoring fluctuation factors; iterating through the sequence of Q lifting point monitoring data groups and the sequence of K steel structure stress point monitoring groups to perform maximum amplitude ratio analysis to obtain Q second lifting point sub-monitoring fluctuation factors and K second steel structure stress point sub-monitoring fluctuation factors; performing weighted calculation on the Q first lifting point sub-monitoring fluctuation factors and the Q second lifting point sub-monitoring fluctuation factors, and the K first steel structure stress point sub-monitoring fluctuation factors and the K second steel structure stress point sub-monitoring fluctuation factors respectively to obtain Q lifting point monitoring fluctuation factors and K steel structure stress point monitoring fluctuation factors; iterating through the sequence of Q lifting point monitoring data groups and the sequence of K steel structure stress point monitoring groups to identify data feature iterative correlation to obtain Q lifting point monitoring correlation factors and K steel structure stress point monitoring correlation factors, comprising: extracting a first hoisting point monitoring data group sequence from the Q hoisting point monitoring data group sequences; iteratively associating and identifying data features of adjacent two first hoisting point monitoring features in the first hoisting point monitoring feature sequence to obtain a first hoisting point monitoring feature correlation factor sequence; calculating a mean value of the first hoisting point monitoring feature correlation factor sequence to obtain a first hoisting point monitoring correlation factor; iteratively associating and identifying data features of adjacent two first hoisting point monitoring features in the first hoisting point monitoring feature sequence to obtain a first hoisting point monitoring feature correlation factor sequence. extracting any adjacent two first hoisting point monitoring features in the first hoisting point monitoring feature sequence; 2. The method for safety warning of roof pipe truss steel structure hoisting under sensing monitoring according to claim 1, characterized in that, performing feature similarity identification on the two first hoisting point monitoring features to obtain a first hoisting point monitoring feature similarity set; performing normalization processing on the first hoisting point monitoring feature similarity set and filling it into an initially empty matrix to obtain a first correlation matrix; performing correlation factor identification based on the first correlation matrix and the two first hoisting point monitoring features to obtain a first hoisting point monitoring feature correlation factor; performing correlation factor identification based on the first correlation matrix and the two first hoisting point monitoring features to obtain a first hoisting point monitoring feature correlation factor, including: performing convolution operation on the first correlation matrix and the two first hoisting point monitoring features respectively to obtain two first hoisting point monitoring enhanced features; performing correlation factor identification based on the first correlation matrix and the two first hoisting point monitoring features to obtain a first hoisting point monitoring feature correlation factor, including:

3. The method for safety warning of roof pipe truss steel structure hoisting under sensing monitoring according to claim 2, characterized in that, obtaining a first safety early warning information by asynchronously double-channel safety monitoring on the first-level risk monitoring object and the second-level risk monitoring object, including: obtaining a historical hoisting log set; iteratively associating and identifying data features of adjacent two first hoisting point monitoring features in the first hoisting point monitoring feature sequence to obtain a first hoisting point monitoring feature correlation factor sequence.

4. The method for safety warning of roof pipe truss steel structure hoisting under sensing monitoring according to claim 1, characterized in that, performing risk classification on the historical hoisting log set to obtain a first-level historical hoisting log set and a second-level historical hoisting log set; performing abnormal time window identification on the first-level historical hoisting log set and the second-level historical hoisting log set to obtain a first-level abnormal time window set and a second-level abnormal time window set; performing monitoring frequency configuration based on the sizes of the first-level abnormal time window set and the second-level abnormal time window set to obtain a first-level monitoring frequency and a second-level monitoring frequency; constructing the first monitoring sub-channel and the second monitoring sub-channel of the asynchronous double-channel according to the first-level monitoring frequency and the second-level monitoring frequency. presetting a warning feedback window, performing abnormal troubleshooting in the warning feedback window, and obtaining a first warning feedback information. ​ 5. The method for safety warning of roof pipe truss steel structure hoisting under sensing monitoring according to claim 1, characterized in that, ​ 6. The method for safety warning of roof pipe truss steel structure hoisting under sensing monitoring according to claim 1, characterized in that, The sensor monitoring array includes strain sensors, displacement sensors, load sensors, and accelerometers.

7. The roof pipe truss steel structure hoisting safety early warning device under the sensing monitoring, characterized in that, The device comprises: a key point acquisition module, configured to acquire Q lifting points and K steel structure stress points of a target pipe truss steel structure, wherein Q and K are positive integers; a data acquisition module, configured to continuously monitor the Q lifting points and the K steel structure stress points by using a sensor monitoring array to obtain Q lifting point monitoring data group sequences and K steel structure stress point monitoring group sequences; a risk identification module, configured to identify risk coefficients by traversing the Q lifting point monitoring data group sequences and the K steel structure stress point monitoring group sequences to determine first-level risk monitoring objects and second-level risk monitoring objects; a safety monitoring module, configured to perform asynchronous dual-channel safety monitoring on the first-level risk monitoring objects and the second-level risk monitoring objects to obtain first safety warning information, wherein the asynchronous dual-channel includes a first monitoring sub-channel and a second monitoring sub-channel.

Images