Intelligent and safe hoisting monitoring method for large-tonnage I-beam

Abstract

The invention discloses an intelligent and safe hoisting monitoring method for a large-tonnage I-beam, relates to the field of construction safety, is based on a monitoring system consisting of a sensing layer, a transmission layer, a data layer, a decision layer and an execution layer, and comprises the following steps: deploying the sensing layer, constructing the transmission layer, configuring the data layer, constructing the decision layer and responding to the execution layer. Structural stress is monitored through a high-precision strain gauge and a distributed optical fiber sensor, the posture is monitored through an MEMS tilt angle sensor and a six-axis gyroscope, the state of the lifting appliance is monitored through a tension sensor and a visual recognition sensor, the environment is monitored through an ultrasonic anemograph and a laser radar, and multi-dimensional data collection is conducted through a displacement sensor monitoring device. Full-scene monitoring of the state of the I-shaped beam, the performance of a lifting appliance, the working environment and crane equipment is achieved, potential risks are captured without dead corners, the accuracy and reliability of data under complex working conditions are ensured, the problems that traditional monitoring data are prone to being interfered by the environment and insufficient in precision are solved, and the safety coefficient is increased.

Description

Technical Field

[0001] This invention relates to the field of construction safety technology, specifically to an intelligent safety monitoring method for the hoisting of large-tonnage I-beams. Background Technology

[0002] Large-tonnage I-beams, as key components of heavy engineering structures, are widely used in bridge construction, large factory building construction, and heavy equipment installation. Their hoisting operations are characterized by large loads (hundreds to thousands of tons), complex operating environments (such as open-air sites, multiple equipment operations, and extreme weather), and numerous risk points (structural stress exceeding limits, posture imbalance, lifting equipment failure, environmental interference, etc.). Once a safety accident occurs, it may lead to serious consequences such as deformation and damage to the I-beam, equipment overturning, and personal injury or death. Therefore, extremely high requirements are placed on real-time safety monitoring of the hoisting process.

[0003] Existing hoisting monitoring technologies have many limitations: traditional methods rely heavily on manual inspections and experience-based judgments, making it difficult to accurately capture key parameters such as dynamic stress and minute posture changes, resulting in poor real-time performance; sensor applications are mostly single-dimensional monitoring (such as monitoring only stress or lifting tension), lacking collaborative perception of structure, lifting equipment, environment, and equipment, which can easily lead to missed risk assessments. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent safety monitoring method for the hoisting of large-tonnage I-beams, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for intelligent safety monitoring of large-tonnage I-beam hoisting, based on a monitoring system composed of a perception layer, a transmission layer, a data layer, a decision layer, and an execution layer, the steps of which include:

[0006] Step 1: Deploy the perception layer to comprehensively capture the stress parameters of the I-beam structure, the attitude parameters of the I-beam, the status parameters of the lifting equipment, and the environmental and equipment parameters during the hoisting process through a distributed sensor network;

[0007] Step 2: Build the transmission layer and adopt a hybrid transmission solution with wireless as the primary method and wired backup for the complex environment of the construction site.

[0008] Step 3: Configure the data layer, filter the parameter data, and use data normalization technology to convert data of different magnitudes and units into a unified standard format.

[0009] Step 4: Construct a decision-making layer by establishing a security threshold database, formulating real-time criteria, and using AI-assisted decision-making for analysis and early warning.

[0010] Step 5: Execution layer response. The parameter changes are displayed through human-computer interaction technology, and corresponding operations are triggered based on the warning level to carry out hierarchical device linkage.

[0011] Preferably, in step one, the stress parameters of the I-beam structure are monitored using high-precision strain gauges and distributed fiber optic sensors;

[0012] High-precision strain gauges and embedded distributed fiber optic sensors were attached to the welds connecting the flanges and webs of the I-beam, the upper and lower flanges at the hoisting points, and the easily flexible parts at mid-span.

[0013] Preferably, in step one, the attitude parameters of the I-beam are monitored using a MEMS tilt sensor and a six-axis gyroscope;

[0014] Both the MEMS tilt sensor and the six-axis gyroscope are installed at the center and both ends of the top surface of the I-beam, respectively, to monitor the overall attitude and to monitor torsional differences.

[0015] Preferably, in step one, multiple tension sensors are connected in series at the connection point between the spreader and the I-beam, and a vision sensor is fixed at a key part of the hook, so that both can monitor the status parameters of the spreader simultaneously.

[0016] Preferably, in step one, an ultrasonic anemometer and a lidar are installed on the top of the crane boom to monitor the wind speed at the work site and detect the distance and position of surrounding obstacles, respectively.

[0017] Displacement sensors are installed at the bottom of the outriggers of the crane used for hoisting operations to monitor vertical settlement, horizontal offset, and displacement differences between outriggers.

[0018] Preferably, in step two, the hybrid transmission scheme is as follows:

[0019] Short range: Long range WAN or industrial-grade Bluetooth;

[0020] Long distance: 5G;

[0021] Key data: redundant transmission.

[0022] Preferably, in step three, the data layer is divided into:

[0023] Edge computing nodes: deployed in the crane operator's cab to preprocess raw data;

[0024] Cloud / local server: Stores historical data for traceability and operational condition analysis, and runs high-precision models.

[0025] Preferably, in step four, the safety threshold library dynamically adjusts the safety range of each parameter according to the I-beam specifications;

[0026] Real-time criteria include single parameter exceeding the limit, multiple parameter exceeding the limit, and trend warning;

[0027] AI-assisted decision-making uses historical hoisting data to train models, identify risks, and intervene in advance.

[0028] Preferably, in step five, the linkage of the grading equipment specifically includes:

[0029] Level 1 warning: An audible and visual alarm alerts the operator.

[0030] A level-two warning is issued, triggering automatic speed reduction for the crane.

[0031] Level 3 alert, hoisting operations suspended;

[0032] In an emergency, immediately shut down the machine and cut off the power source.

[0033] Compared with the prior art, the beneficial effects of the present invention are:

[0034] 1. This invention breaks through the limitations of traditional single-dimensional monitoring. It uses high-precision strain gauges and distributed fiber optic sensors to monitor structural stress, MEMS tilt sensors and six-axis gyroscopes to monitor attitude, tension sensors and visual recognition sensors to monitor the status of the lifting equipment, ultrasonic anemometers and lidar to monitor the environment, and displacement sensors to monitor equipment. This multi-dimensional data collection enables full-scene monitoring of the I-beam's own status, lifting equipment performance, operating environment, and crane equipment. It captures potential risks such as stress mutations, attitude imbalances, lifting equipment failures, and environmental interference without blind spots, ensuring the accuracy and reliability of data under complex working conditions and solving the problems of traditional monitoring data being easily affected by environmental interference and lacking accuracy.

[0035] 2. In this invention, an AI algorithm trained on a dynamically adjusted preset safety threshold library, single parameter exceeding the limit, multi-parameter linkage and trend warning multi-dimensional criteria, and historical data is used to achieve full-level decision-making from single parameter exceeding the limit to multi-factor coupled risk, from instant alarm to trend prediction. This solves the problem of traditional methods that rely solely on simple threshold comparison and miss hidden risks, making the warning more accurate and comprehensive.

[0036] 3. In this invention, the execution layer forms a closed loop of "monitoring-analysis-decision-execution" through human-computer interaction and hierarchical early warning response, replacing the traditional mode that relies on manual response, greatly improving the speed and accuracy of emergency response, and minimizing the risk of accidents. Attached Figure Description

[0037] Figure 1 This is a flowchart of a method for monitoring the intelligent and safe hoisting of large-tonnage I-beams according to the present invention. Detailed Implementation

[0038] 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.

[0039] Example: Refer to Figure 1 As shown: A smart safety hoisting monitoring method for large-tonnage I-beams is based on a monitoring system composed of a perception layer, transmission layer, data layer, decision layer, and execution layer. The system is as follows:

[0040] 1. Perception Layer: Multi-dimensional Data Acquisition Module

[0041] Through a distributed sensor network, dynamic parameters of the hoisting process are comprehensively captured. The core sensing unit includes:

[0042]

[0043]

[0044] The specific sensors are as follows:

[0045] The high-precision strain gauge, in addition to an accuracy of ±1με, also exhibits excellent temperature stability, with a temperature drift of ≤0.5με / ℃ within the range of -20℃ to 60℃, ensuring the accuracy of measurement results even when there are significant temperature variations at the hoisting operation site. Its response frequency is ≥1kHz, enabling it to promptly capture dynamic stress changes in the I-beam during the hoisting process.

[0046] Its installation location is as follows:

[0047] (1) Flange-web connection weld (critical stress concentration node): Specifically, at the three high-risk sections of the longitudinal connection weld—the "mid-span section," the "1 / 4 span section," and the "section corresponding to the lifting point"—two sets of strain gauges are placed at each section (one set for the upper flange-web weld and one set for the lower flange-web weld). Each set contains two strain gauges, which are respectively attached 5mm from the weld edge (the area with the highest stress concentration coefficient) and on the weld centerline (mean stress reference) to form a "two-point comparison" monitoring.

[0048] Monitoring objective: Welds are weak points in the structure. The shear stress and tensile and compressive stress of welds at the mid-span, 1 / 4 span and suspension points are the greatest. Strain gauges can accurately capture the stress peaks of these key sections (for example, the shear stress of welds at suspension points may reach more than 80% of the allowable value of the material), making up for the insufficient resolution of fiber optic sensors in measuring local peak stress.

[0049] (2) Near the hoisting point (the stress peak area of ​​the upper and lower flanges), the specific location is: on the upper and lower flange surfaces directly below each hoisting point (within ±300mm from the center of the hoisting point), 4 strain gauges are arranged (2 on the upper flange and 2 on the lower flange).

[0050] Upper flange (tension zone): A strain gauge is placed 50mm directly below the lifting point (to measure the peak value of the tensile stress directly applied to the lifting point), and another strain gauge is placed 100mm outside the lifting point (to measure the stress gradient at the edge of the lifting point).

[0051] Lower flange (compression zone): Arranged symmetrically with the upper flange, measuring the peak compressive stress and distribution gradient below the suspension point.

[0052] Monitoring objective: The lifting point is the core of force transmission, and the stress concentration factor can reach 2 to 3 times. Strain gauges can accurately measure the instantaneous stress peak near the lifting point (such as the impact stress when lifting starts), providing direct data for load matching of the lifting equipment (which needs to be cross-validated with the data from the lifting equipment tension sensor).

[0053] (3) The most flexible part in the middle of the span (the point of maximum stress) is located at the center of the upper flange, the center of the lower flange, and the center of the web of the I-beam, forming a "triangular monitoring group".

[0054] Center of upper flange: Measure the maximum tensile stress at mid-span (the upper flange experiences the most significant tension during deflection);

[0055] Lower flange center: Measure the maximum compressive stress at mid-span (the core point prone to buckling);

[0056] Web center: Measure the maximum shear stress at mid-span (the concentrated area of ​​shear deformation in the web during deflection).

[0057] Monitoring objective: The mid-span is the area with the largest flexural deformation of the I-beam, and the stress value is close to the design limit. The strain gauge can capture the stress peak at this point in real time (accuracy up to ±0.5MPa) and directly determine whether it is close to the material yield strength (e.g., the yield strength of Q355 steel is 355MPa, and an early warning is triggered when the stress reaches 310MPa).

[0058] Distributed fiber optic sensors (BOTDR / BOTDA): With a spatial resolution of up to 1m and a measurement range of 0–10000με, they enable distributed stress monitoring of I-beams over distances of several kilometers, providing a comprehensive understanding of the stress distribution within the I-beams. Furthermore, they possess characteristics such as electromagnetic interference resistance, corrosion resistance, and high-pressure resistance, making them suitable for use in harsh hoisting environments.

[0059] Its installation location is as follows:

[0060] (1) Flange and web connection weld (continuous coverage along the entire length), specific location: along the entire length of the longitudinal connection weld between the upper flange and web and the lower flange and web, one optical fiber is arranged in parallel (50mm from the edge of the weld) to cover the complete weld area from the beam end to the beam end.

[0061] Monitoring objective: Potential defects exist along the entire length of the weld (such as local lack of fusion or porosity). Fiber optics can continuously monitor the strain distribution along the entire length of the weld, capturing "local strain spikes" in areas not covered by strain gauges (such as a weld strain suddenly exceeding 20% ​​of the surrounding area, which may be a precursor to defect expansion), thus filling the blind spots of strain gauge "point monitoring".

[0062] (2) Near the hoisting point (upper and lower flanges, within 3m of the hoisting point), the specific location is: with the hoisting point of the upper and lower flanges as the center, extend 1.5m to both sides along the beam length direction and arrange 1 optical fiber (1 on the upper flange and 1 on the lower flange) to form the "core monitoring zone of the hoisting point".

[0063] Monitoring objective: The stress gradient near the suspension point is large (the peak stress at the suspension point decreases rapidly to both sides). The optical fiber can continuously capture this gradient change to determine whether the stress at the suspension point is uniform (e.g., if the strain difference at 1m on both sides exceeds 10%, there may be a skew in the suspension). At the same time, it can monitor whether the stress transmission path between the suspension point and the mid-span is abnormal.

[0064] (3) The easily flexed part in the middle of the span (continuous coverage of the 10m range in the middle of the span), the specific location is: extend 5m to both sides of the middle section (a total range of 10m), and arrange one optical fiber on the upper flange, lower flange and web respectively (overlapping with the "triangular monitoring group" of the strain gauge) to form a "three-dimensional continuous monitoring area".

[0065] Monitoring objective: The 10m mid-span is the area with the most significant flexural deformation. Fiber optics can continuously monitor the strain distribution curve within this range (one data point every 0.5m) to identify whether there are local strain anomalies (such as a sudden strain at a certain point exceeding the surrounding average by 15%, which may be a material defect or local damage), and provide distribution verification for the point data of the strain gauge.

[0066] MEMS tilt sensor: In addition to an accuracy of ±0.1°, the measurement range is ±30°, which can meet the monitoring requirements of large angle changes of I-beams during hoisting. Its output frequency is 100Hz, which can reflect the attitude changes of the I-beam in real time.

[0067] Six-axis gyroscope: dynamic response ≥100Hz, measurement range ±2000° / s (angular velocity) and ±16g (acceleration), can accurately measure the rotational angular velocity and linear acceleration of the I-beam, providing more comprehensive data for attitude monitoring.

[0068] Tension sensor: Range matches the rated load of the lifting device, accuracy ±1% FS, nonlinearity error ≤0.5% FS, hysteresis error ≤0.5% FS, accurately measuring the tension of the lifting device. It has good impact resistance, still functioning normally when the impact load is 150% of the rated load.

[0069] Visual sensor: High-definition camera with a resolution of ≥1920×1080 and a frame rate of ≥30fps. Combined with a specialized crack recognition algorithm, it can clearly capture images of key parts of the hook and quickly identify defects such as cracks, with an accuracy rate of ≥95%.

[0070] Ultrasonic anemometer: Measurement range 0~60m / s, accuracy ±0.5m / s, resolution 0.1m / s, can monitor the wind speed at the work site in real time, and provide wind load reference for hoisting operations.

[0071] LiDAR: Scanning radius ≥20m, angular resolution ≤0.1°, capable of accurately detecting the distance and position of surrounding obstacles, measurement accuracy ±5cm, ensuring a safe distance between the I-beam and obstacles.

[0072] 2. Transmission Layer: High-reliability data link, employing a hybrid transmission scheme of "wireless as the primary method and wired backup" to address the complex electromagnetic environment of construction sites.

[0073] (1) Short distance (from sensor to field edge node): LoRaWAN (long distance wide area network, low power consumption, anti-interference, transmission distance ≥500m) or industrial Bluetooth (suitable for short distance high frequency data, such as attitude sensor);

[0074] (2) Long distance (from edge node to monitoring center): 5G (high bandwidth, low latency, supports real-time video and massive sensor data, latency ≤100ms);

[0075] (3) Key data (such as stress, spreader tension): redundant transmission (dual path transmission) to avoid data loss.

[0076] It uses an industrial-grade wireless gateway with an operating temperature range of -40℃ to 85℃, making it suitable for extreme temperature environments at hoisting sites; it also has vibration resistance, with an acceleration of up to 50m / s² at vibration frequencies of 10Hz to 2000Hz. 2 It can withstand vibration and impact during hoisting; at the same time, it has strong anti-electromagnetic interference capabilities, which can effectively resist electromagnetic interference generated by cranes and other equipment, ensuring the real-time and reliability of data transmission, with a transmission rate of over 100Mbps, meeting the system's large data transmission needs.

[0077] 3. Data layer: Real-time processing and storage.

[0078] (1) Edge computing nodes: Deployed near the crane operator’s cab to preprocess the raw data (filtering, noise reduction, outlier removal), for example, by optimizing attitude data through Kalman filtering to reduce vibration interference;

[0079] (2) Cloud / local server: Stores historical data (for tracing and working condition analysis) and runs high-precision models (such as finite element real-time stress inversion model, which calculates the overall stress distribution based on the stress at monitoring points).

[0080] Utilizing high-performance industrial servers, the CPU features a multi-core processor with a clock speed of ≥3.0GHz, ≥32GB of memory, and ≥2TB SSD storage, enabling rapid processing and storage of large amounts of data collected by the system. The server incorporates a redundant power supply design to ensure continued operation and prevent data loss in the event of a sudden power outage. Furthermore, it supports virtualization technology, allowing for flexible resource allocation based on system requirements, thereby improving server utilization.

[0081] The data layer employs a Kalman filter algorithm to filter the acquired attitude data, effectively removing noise interference and improving data smoothness and accuracy. For stress data, a wavelet transform algorithm is used for noise reduction, which better preserves the abrupt changes in the signal and accurately captures instantaneous stress variations. Furthermore, data normalization technology is used to convert data of different magnitudes and units into a unified standard format, facilitating subsequent data analysis and comparison.

[0082] 4. Decision-making layer: Intelligent analysis and early warning logic, based on rule engine and AI algorithm to realize multi-level decision-making.

[0083] (1) Safety threshold library: preset safety ranges for each parameter (such as stress of Q345 steel I-beam ≤235MPa, tilt angle ≤5°, wind speed ≤8m / s, etc., dynamically adjusted according to the I-beam specifications);

[0084] (2) Real-time criteria, including:

[0085] Single parameter exceeding the standard: such as stress > 235MPa → Level 1 warning (audio and visual alarm + screen flashing);

[0086] Multi-parameter linkage: such as tilt angle > 5° + tension of one side of the spreader > 1.2 times the average value → Level 2 warning (forced deceleration + voice prompt adjustment);

[0087] Trend warning: If the stress increases by ≥50MPa within 5 minutes (not exceeding the standard but the rate of increase is too fast) → early warning (suggest checking the lifting point);

[0088] (3) AI-assisted decision-making: The model is trained by historical hoisting data to identify "hidden risk patterns" (such as a certain type of I-beam is prone to torsion when the wind speed is 6m / s and the tilt is 3°, so early intervention is possible).

[0089] The decision-making team established a mechanical model for I-beam hoisting, combining collected data on stress, posture, and lifting equipment tension to simulate and analyze the stress state of the I-beams. This allowed them to calculate parameters such as stress distribution and deformation at various points on the I-beams, and assess their structural safety. Simultaneously, big data analytics were used to mine historical hoisting data, identifying patterns of risk and potential hazards during the hoisting process, providing decision support for hoisting operations. For example, by analyzing stress changes in the I-beams under different wind speeds and tilt angles, corresponding safe operating thresholds were established.

[0090] 5. Execution layer: Closed-loop control and response.

[0091] (1) Human-computer interaction: stress cloud map, attitude three-dimensional model and lifting tension curve are displayed in real time through the display screen in the control room, and key parameters are highlighted in red.

[0092] (2) Equipment linkage: By interfacing with the crane PLC system, a "early warning-control" closed loop is achieved:

[0093] Level 1 warning → Audible and visual alarm to alert operators;

[0094] Level 2 warning → Crane automatically reduces speed (≤50% of rated speed);

[0095] Level 3 warning → Suspend hoisting (only fine-tuning operations are allowed);

[0096] Emergency situation (such as stress > 300MPa or warning of lifting tool breakage) → Immediate shutdown (cut off power source).

[0097] Based on the set safety thresholds, the system will automatically issue an early warning when the monitored parameters exceed the thresholds. There are three warning levels: Level 1 indicates a slight exceedance, and the system will alert the operator with an audible and visual alarm; Level 2 indicates a moderate exceedance, and in addition to the audible and visual alarm, the system will automatically limit the crane's operating speed to prevent further deterioration of the parameters; Level 3 indicates a more serious exceedance, and the system will immediately issue a stop-lift signal, cutting off the crane's power source in an emergency to immediately stop the machine and ensure operational safety. Simultaneously, the system will send the warning information to the mobile app of relevant management personnel in real time for timely action.

[0098] A method for intelligent safety monitoring during the hoisting of large-tonnage I-beams includes the following steps:

[0099] Step 1: Deploy the perception layer to achieve multi-dimensional data acquisition. This involves comprehensively capturing dynamic parameters of the hoisting process through a distributed sensor network, specifically including:

[0100] Structural stress monitoring: High-precision strain gauges (accuracy ±1με, temperature drift ≤0.5με / ℃ within -20℃~60℃, response frequency ≥1kHz) and distributed fiber optic sensors (BOTDR / BOTDA, spatial resolution 1m, measurement range 0~10000με) are pasted or embedded at the weld joint between the flange and web of the I-beam, near the hoisting point, and in the middle of the span to monitor the strain value and stress distribution of key sections.

[0101] Attitude monitoring: MEMS tilt sensors (accuracy ±0.1°, measurement range ±30°, output frequency 100Hz), six-axis gyroscopes (dynamic response ≥100Hz, measurement range ±2000° / s angular velocity and ±16g acceleration) and GNSS (centimeter-level positioning) are installed at the center and both ends of the top surface of the I-beam to monitor the longitudinal tilt angle (≤5°), lateral tilt angle (≤3°), and torsion angle (≤2°).

[0102] Sling condition monitoring: A tension sensor (accuracy ±1%FS, resistant to 150% rated load impact) is connected in series at the connection point between the sling and the I-beam. A visual sensor (high-definition camera + crack recognition algorithm, accuracy ≥95%) is fixed on the boom to monitor the tension deviation (≤10%) of a single wire rope / sling and the crack / deformation of the hook.

[0103] Environmental and equipment monitoring: An ultrasonic anemometer (range 0~60m / s, accuracy ±0.5m / s) and a lidar (scanning radius ≥20m, accuracy ±5cm) are installed on the top of the crane boom. Displacement sensors are installed on the outrigger bases. Data is read through the crane's CAN bus interface to monitor wind speed (≤10m / s), obstacle distance (≥5m), crane torque (≤90% of rated torque), and outrigger settlement (≤5mm / min).

[0104] Step 2: Establish the transport layer and build a highly reliable data link. A hybrid transmission scheme of "wireless as the primary method and wired backup" is adopted to address the complex environment of the construction site.

[0105] Short-range transmission (sensor to edge node): LoRaWAN (transmission distance ≥ 500m) or industrial-grade Bluetooth (suitable for high-frequency data such as attitude sensors);

[0106] Long-distance transmission (from edge node to monitoring center): High-bandwidth, low-latency transmission is achieved through 5G network (latency ≤100ms), supporting real-time video and massive sensor data;

[0107] Critical data transmission: Dual-path redundant transmission is used for critical data such as stress and spreader tension to avoid loss;

[0108] Configured with an industrial-grade wireless gateway (operating temperature -40℃~85℃, 50m / s² acceleration under vibration at 10Hz~2000Hz). 2 It is resistant to electromagnetic interference and has a transmission rate of ≥100Mbps, ensuring the real-time performance and reliability of data transmission in extreme environments.

[0109] Step 3: Configure the data layer to complete real-time processing and storage.

[0110] Edge computing node preprocessing: Edge nodes are deployed near the crane operator's cab. Kalman filtering is used to optimize attitude data, wavelet transform is used to reduce stress data noise, outliers are removed, and signal mutation characteristics are preserved.

[0111] Cloud / local server storage and modeling: An industrial server (with redundant power supply) with a multi-core processor (clock frequency ≥ 3.0 GHz), 32 GB or more of memory, and 2 TB SSD or more of hard drive is used to store historical data and run a finite element real-time stress inversion model to calculate the overall distribution based on the stress at monitoring points;

[0112] Data standardization processing: Normalization technology is used to convert data of different magnitudes and units into a unified format, and virtualization technology is used to flexibly allocate server resources.

[0113] Step 4: Construct a decision-making layer to achieve intelligent analysis and early warning.

[0114] Establish a safety threshold library: preset safety ranges for parameters, such as stress of Q345 steel I-beams ≤235MPa, wind speed ≤8m / s, crane torque ≤90% of rated torque, etc. (dynamically adjusted according to I-beam specifications);

[0115] Establish real-time judgment criteria: a single parameter exceeding the standard (such as stress > 235MPa) triggers a first-level warning (audio-visual alarm + screen flashing); multiple parameter linkage (such as tilt angle > 5° + tension of one side of the lifting device > 1.2 times the average value) triggers a second-level warning (forced deceleration + voice prompt); abnormal trends (such as stress rise ≥ 50MPa within 5 minutes) trigger an early warning (prompt to check the lifting point);

[0116] AI-assisted decision-making: Based on historical data, the model is trained to identify hidden risk patterns (such as easy to twist when the wind speed is 6m / s and the tilt is 3°), and the structural safety is evaluated by simulating the mechanical model.

[0117] Step 5: Execution layer response, forming closed-loop control.

[0118] Human-computer interaction demonstration: The stress cloud map, 3D posture model and tension curve of the spreader are displayed in real time on the display screen in the control room, and key parameters are highlighted in red;

[0119] Tiered equipment linkage: Level 1 warning only alerts the operator and pushes information to the management personnel's APP; Level 2 warning automatically reduces the crane speed to ≤50% of the rated speed; Level 3 warning suspends lifting (only fine-tuning is allowed); in emergency situations (such as stress >300MPa or lifting tool breakage warning), the power source is immediately cut off to stop the machine and ensure operational safety.

[0120] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent safety monitoring during the hoisting of large-tonnage I-beams, characterized in that, A monitoring system consisting of a perception layer, a transmission layer, a data layer, a decision layer, and an execution layer includes the following steps: Step 1: Deploy the perception layer to comprehensively capture the stress parameters of the I-beam structure, the attitude parameters of the I-beam, the status parameters of the lifting equipment, and the environmental and equipment parameters during the hoisting process through a distributed sensor network; Step 2: Build the transmission layer and adopt a hybrid transmission solution with wireless as the primary method and wired backup for the complex environment of the construction site. Step 3: Configure the data layer, filter the parameter data, and use data normalization technology to convert data of different magnitudes and units into a unified standard format. Step 4: Construct a decision-making layer by establishing a security threshold database, formulating real-time criteria, and using AI-assisted decision-making for analysis and early warning. Step 5: Execution layer response. The parameter changes are displayed through human-computer interaction technology, and corresponding operations are triggered based on the warning level to carry out hierarchical device linkage.

2. The intelligent safety hoisting monitoring method for large-tonnage I-beams according to claim 1, characterized in that, In step one, the stress parameters of the I-beam structure are monitored using high-precision strain gauges and distributed fiber optic sensors; High-precision strain gauges and embedded distributed fiber optic sensors were attached to the welds connecting the flanges and webs of the I-beam, the upper and lower flanges at the hoisting points, and the easily flexible parts at mid-span.

3. The intelligent safety hoisting monitoring method for large-tonnage I-beams according to claim 1, characterized in that, In step one, the attitude parameters of the I-beam are monitored using a MEMS tilt sensor and a six-axis gyroscope; Both the MEMS tilt sensor and the six-axis gyroscope are installed at the center and both ends of the top surface of the I-beam, respectively, to monitor the overall attitude and to monitor torsional differences.

4. The intelligent safety hoisting monitoring method for large-tonnage I-beams according to claim 1, characterized in that, In step one, multiple tension sensors are connected in series at the connection point between the spreader and the I-beam, and a vision sensor is fixed at a key part of the hook. Both monitor the status parameters of the spreader simultaneously.

5. The intelligent safety hoisting monitoring method for large-tonnage I-beams according to claim 1, characterized in that, In step one, an ultrasonic anemometer and a lidar are installed on the top of the crane boom to monitor the wind speed at the work site and detect the distance and position of surrounding obstacles, respectively. Displacement sensors are installed at the bottom of the outriggers of the crane used for hoisting operations to monitor vertical settlement, horizontal offset, and displacement differences between outriggers.

6. The intelligent safety hoisting monitoring method for large-tonnage I-beams according to claim 1, characterized in that, In step two, the hybrid transmission scheme is as follows: Short range: Long range WAN or industrial-grade Bluetooth; Long distance: 5G; Key data: redundant transmission.

7. The intelligent safety hoisting monitoring method for large-tonnage I-beams according to claim 1, characterized in that, In step three, the data layer is divided into: Edge computing nodes: deployed in the crane operator's cab to preprocess raw data; Cloud / local server: Stores historical data for traceability and operational condition analysis, and runs high-precision models.

8. The intelligent safety hoisting monitoring method for large-tonnage I-beams according to claim 1, characterized in that, In step four, the safety threshold library dynamically adjusts the safety range of each parameter according to the I-beam specifications; Real-time criteria include single parameter exceeding the limit, multiple parameter exceeding the limit, and trend warning; AI-assisted decision-making uses historical hoisting data to train models, identify risks, and intervene in advance.

9. The intelligent safety hoisting monitoring method for large-tonnage I-beams according to claim 1, characterized in that, In step five, the linkage of the grading equipment specifically refers to: Level 1 warning: An audible and visual alarm alerts the operator. A level-two warning is issued, triggering automatic speed reduction for the crane. Level 3 alert, hoisting operations suspended; In an emergency, immediately shut down the machine and cut off the power source.

Images