Method for monitoring the safety risks of the operating state of a bridge-launching machine

Abstract

The application discloses a kind of working state safety risk monitoring methods for bridge erecting machine, the method includes: based on bridge erecting machine operation safety perception, real-time sensing data deployed on bridge erecting machine is collected, and is uploaded to local edge node and cloud platform;Build digital twin model, the real-time sensing data collected is mapped to virtual model, realize operation state visual presentation;Using random forest algorithm, the historical operation failure data collected is trained, and risk identification model is obtained, and the data collected in real time is identified based on risk identification model Operation risk;Structural stress, attitude instability and personnel safety distance boundary early warning are constructed, and the data collected in real time are risk identified;Different risk levels are executed predetermined bridge erecting machine operation early warning and safety control based on, solve the problem of artificial dependence, response lag of bridge erecting machine safety control, realize from passive prevention and control to active early warning, accurate control, guarantee operation safety, give consideration to construction efficiency.

Description

Technical Field

[0001] This invention relates to the field of bridge erecting machine safety operation monitoring technology, and more specifically to a method for monitoring the safety risks of bridge erecting machine operation status. Background Technology

[0002] As a core heavy equipment in bridge construction, bridge erecting machines often operate at heights, under heavy loads, and in dynamic conditions. They are also significantly affected by environmental factors such as wind speed and visibility. Therefore, safety management is directly related to the safety of construction personnel and the progress of the project. Current traditional safety management methods for bridge erecting machines rely on manual inspections and experience-based judgment. Hidden risks such as changes in internal stress of the main beam and minute deviations in the lifting equipment's posture are often undetectable to the naked eye and only become apparent before an accident occurs. This approach is not only inefficient and prone to missed detections, making it difficult to capture hidden risks such as structural stress changes and posture deviations in real time, but also lacks comprehensive coverage of personnel location and environmental conditions, leading to incomplete risk assessments.

[0003] In control measures based on on-site data collection, the control standards are quantified by experience or by the safety supervisor during inspections. The response methods are singular, which can easily lead to imbalances such as excessive downtime affecting efficiency or high-risk underreporting, making it difficult to meet the safety requirements of modern bridge construction. Summary of the Invention

[0004] The purpose of this invention is to provide a method for monitoring the safety risks of bridge erecting machines in operation. With multi-dimensional perception, intelligent analysis and closed-loop management as the core architecture, it overcomes the problems of manual dependence and slow response in the traditional safety management of bridge erecting machines, and realizes the transformation from passive prevention and control to active early warning and precise management, so as to ensure operational safety and take into account construction efficiency.

[0005] According to a first aspect of the present invention, a method for monitoring the operational safety risks of a bridge erecting machine is provided, comprising the following steps: Step 1: Based on the safety perception of the bridge erecting machine, collect real-time sensing data from the structural stress sensor, working posture sensor, working environment sensor and personnel positioning sensor deployed on the bridge erecting machine, and upload them to the local edge node and cloud platform. Step 2: Based on the bridge erecting machine design drawings and BIM model, construct a 1:1 digital twin model, and map the collected real-time sensor data to the virtual model to realize the visualization of the operation status; Step 3: Use the random forest algorithm to train the risk identification model using historical operation failure data, and use the risk identification model to identify operation risks based on the real-time collected data, outputting operation risk levels 0-3, where level 0 is no risk, level 1 is low risk, level 2 is high risk, and level 3 is extremely high risk. Step 4: Construct early warning systems for structural stress, attitude instability, and personnel safety distance boundaries. Perform risk identification on the real-time collected data and output risk levels from 0 to 3, where level 0 is no risk, level 1 is low risk, level 2 is high risk, and level 3 is extremely high risk. Step 5: Based on the risk level output from boundary warning and operational risk identification, execute the predetermined bridge erecting machine operation warning and safety control.

[0006] As an optional implementation, in step 1, the structural stress sensor is a fiber optic grating type sensor, installed at the mid-span of the main beam and at the connection of the support legs, with a sampling frequency of 10Hz. The working attitude sensor adopts an IMU inertial measurement unit, which is installed on the crane trolley and the lifting device to monitor the pitch angle and roll angle, with a sampling frequency of 10Hz; The environmental sensor is deployed at a high point in the work area to monitor wind speed, visibility, and temperature, with a sampling frequency of 10Hz; The personnel positioning sensor uses UWB positioning and equips the operators with positioning tags to capture the distance between the personnel and the dangerous area of ​​the bridge erecting machine in real time. The dangerous area of ​​the bridge erecting machine includes the area under the outriggers and the area covered by the spreader.

[0007] As an optional implementation, the real-time sensing data uses GPS timing technology to synchronize the timestamps of all sensor data, synchronizing the sensor timestamps every 10 seconds. A time synchronization server is deployed on the local edge node, and a unified timestamp is added to all sensor data based on the NTP protocol to ensure that the time synchronization error of multi-source data is ≤1ms.

[0008] As an optional implementation, in step 2, a 1:1 digital twin model is constructed based on the bridge erecting machine design drawings and BIM model. The collected real-time sensor data is mapped to the virtual model to achieve a visual presentation of the operation status, including: Step 2-1: Obtain the CAD design drawings, Revit BIM model, equipment rated load, structural dimensions, material performance parameters, and terrain data of the working scene for the bridge erecting machine; Step 2-2: Import the BIM model using Unity 3D software and recreate the detailed structure of the bridge erecting machine at a 1:1 scale, including the main beam, legs, lifting trolley, and lifting gear, with material parameters consistent with the actual structure. Steps 2-3: Build a work scenario model, including bridge piers, construction access roads, and surrounding environment; import terrain and elevation data to ensure that the scenario matches the actual working conditions. Steps 2-4: Real-time data mapping: Through the data interface API, real-time data transmission between edge nodes and the digital twin platform is realized, and the collected real-time sensor data is mapped to the corresponding positions of the model: stress data is presented as a color cloud map, blue ≤50MPa, green 50-100MPa, yellow 100-150MPa, red ≥150MPa, attitude data drives the model to move synchronously, personnel positions are marked with dynamic icons, and environmental data is displayed with numerical labels superimposed.

[0009] As an optional implementation, in step 3, a random forest algorithm is used to train a risk identification model using collected historical operation failure data. Based on this model, operational risks are identified from the real-time collected data, including: Step 3-1: Collect historical data of bridge erecting machine operation, including normal operation data and fault data. The fault data includes structural fatigue, attitude instability and personnel violation fault types. Divide the historical dataset into training set, validation set and test set in a ratio of 7:2:1. Step 3-2: Select 8 key features: maximum stress of main beam, outrigger stress, pitch angle of lifting equipment, roll angle, wind speed, visibility, distance between personnel and danger zone, and operating load. Use the Pearson correlation coefficient method to eliminate redundant features; then standardize the selected features based on the Z-score formula. Step 3-3: Build a random forest model based on the Scikit-learn library. Initial parameter settings: number of decision trees 100, maximum tree depth 10, minimum number of sample splits 2. Then, the model is trained using the aforementioned training set. The accuracy is tested with the validation set every 10 iterations. The parameters are tuned using the grid search method. Training is stopped when the accuracy on the validation set is ≥95%. The model was validated using the aforementioned test set, with an accuracy of ≥92% and a recall of ≥90%. If the accuracy was not met, the model parameters were readjusted and the model was retrained until the requirements were met. Steps 3-4: Deploy the trained risk identification model on local edge nodes to identify operational risks from real-time collected data and output operational risk levels of 0-3, where level 0 is no risk, level 1 is low risk, level 2 is high risk, and level 3 is extremely high risk.

[0010] As an optional implementation, step 4, which involves constructing early warning systems for structural stress, attitude instability, and personnel safety distance boundaries, includes: Step 4-1: Construct a structural stress boundary early warning system. σ 预警 =[σ 允许 ] / k; Where, σ 允许The allowable stress of the main body material of the bridge erecting machine is 235MPa for Q345 steel and 170MPa for Q235 steel; k is the safety factor, k=2.0 for the main beam area and k=1.8 for the outrigger area; When the real-time stress σ 实时 ≥σ 预警 At that time, a Level 2 warning was triggered; When σ 实时 ≥1.2×σ 预警 At that time, a Level 3 warning was triggered; Step 4-2: Construct an early warning system for attitude instability boundaries: θ 预警 =θ 额定 ×0.8; Where, θ 额定 The bridge erecting machine is designed with a maximum allowable pitch or roll angle; when the real-time attitude angle θ 实时 ≥θ 预警 When θ is triggered, a Level 2 warning is activated; when θ is triggered... 实时 ≥θ 额定 When the value reaches 0.95, a Level 3 warning is triggered.

[0011] Step 4-2: Establish personnel safety distance boundary early warning: The radius of the danger zone is set to R=5m. When the distance between personnel and the boundary of the danger zone is d≤1m, a level 2 warning is triggered; when d≤0.5m, a level 3 warning is triggered.

[0012] The safety risk monitoring method for the operation status of bridge erecting machines described in the above embodiments of the present invention has significant advantages over the prior art as follows: To address the four core risk points of the bridge erecting machine—structure, attitude, environment, and personnel—multiple types of sensors are deployed, and time synchronization is used to provide high-quality data sources for subsequent analysis, thereby fundamentally avoiding safety misjudgments caused by missing information. By leveraging a 1:1 digital twin model to visualize operational status, deploying an AI risk identification model based on random forests, and combining boundary warning quantification to clarify warning thresholds, risk assessment shifts from manual experience to data quantification, achieving a tiered warning mechanism. Different response methods are matched according to risk level, avoiding excessive downtime that affects efficiency while quickly triggering emergency shutdowns in high-risk situations. This solves the problems of vague warning standards and unbalanced responses in traditional warning systems, enabling proactive warning, precise control, and continuous optimization of the bridge erecting machine's operational status, providing full-cycle technical support for bridge construction safety.

[0013] It should be understood that all combinations of the foregoing concepts and the additional concepts described in more detail below may be considered part of the inventive subject matter of this disclosure, provided that such concepts do not contradict each other. Furthermore, all combinations of the claimed subject matter are considered part of the inventive subject matter of this disclosure.

[0014] The foregoing and other aspects, embodiments, and features of the teachings of the present invention will be more fully understood from the following description in conjunction with the accompanying drawings. Other additional aspects of the invention, such as features and / or beneficial effects of exemplary embodiments, will become apparent from the following description or may be learned through practice of specific embodiments according to the teachings of the present invention. Attached Figure Description

[0015] The accompanying drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component shown in the various figures may be denoted by the same reference numeral. For clarity, not every component is labeled in each figure. Embodiments of various aspects of the invention will now be described by way of example and with reference to the accompanying drawings.

[0016] Figure 1 This is a flowchart illustrating a method for monitoring the operational status safety risks of a bridge erecting machine according to an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of the process of constructing a digital twin model and mapping sensor data according to an embodiment of the present invention.

[0018] Figure 3 This is a schematic diagram illustrating the process of risk identification output using a risk identification model according to an embodiment of the present invention. Detailed Implementation

[0019] To better understand the technical content of the present invention, specific embodiments are described below in conjunction with the accompanying drawings.

[0020] Various aspects of the invention are described in this disclosure with reference to the accompanying drawings, in which numerous illustrative embodiments are shown. The embodiments of this disclosure are not necessarily intended to encompass all aspects of the invention. It should be understood that the various concepts and embodiments described above, as well as those described in more detail below, can be implemented in any of many ways, because the concepts and embodiments disclosed herein are not limited to any particular implementation. Furthermore, some aspects of the invention disclosed may be used alone or in any suitable combination with other aspects of the invention disclosed.

[0021] {Example 1} Combination Figure 1 As shown, the method for monitoring the operational status safety risks of a bridge erecting machine according to an embodiment of the present invention includes the following steps: Step 1: Based on the safety perception of the bridge erecting machine, collect real-time sensing data from the structural stress sensor, working posture sensor, working environment sensor and personnel positioning sensor deployed on the bridge erecting machine, and upload them to the local edge node and cloud platform. Step 2: Based on the bridge erecting machine design drawings and BIM model, construct a 1:1 digital twin model, and map the collected real-time sensor data to the virtual model to realize the visualization of the operation status; Step 3: Use the random forest algorithm to train the risk identification model using historical operation failure data, and use the risk identification model to identify operation risks based on the real-time collected data, outputting operation risk levels 0-3, where level 0 is no risk, level 1 is low risk, level 2 is high risk, and level 3 is extremely high risk. Step 4: Construct early warning systems for structural stress, attitude instability, and personnel safety distance boundaries. Perform risk identification on the real-time collected data and output risk levels from 0 to 3, where level 0 is no risk, level 1 is low risk, level 2 is high risk, and level 3 is extremely high risk. Step 5: Based on the risk level output from boundary warning and operational risk identification, execute the predetermined bridge erecting machine operation warning and safety control.

[0022] As an optional embodiment, in step 1, the structural stress sensor is a fiber optic grating type sensor, installed at the mid-span of the main beam and the connection of the outriggers, with a sampling frequency of 10Hz; wherein, according to the structural drawings of the bridge erecting machine, the core stress points such as the mid-span of the main beam and the connection of the outriggers are marked (one stress sensor installation point is set every 2 meters, and two additional ones are added at the upper and lower ends of the outriggers). The working attitude sensor adopts an IMU inertial measurement unit, which is installed on the crane trolley and the spreader to monitor the pitch angle and roll angle, with a sampling frequency of 10Hz. The IMU inertial measurement unit is fixed to the center position of the top of the crane trolley and both ends of the spreader beam by bolts with a bolt torque of 45N·m to ensure that the sensor coordinate system is consistent with the equipment coordinate system. The environmental sensor is deployed at a high point in the work area and fixed with a flange. The levelness error is ≤ ±0.5°. It is used to monitor wind speed, visibility, and temperature, with a sampling frequency of 10Hz. The personnel positioning sensor uses UWB positioning. The operator is equipped with a positioning tag, which is worn on the back of the operator's safety helmet. The terminal is bound to the operator's identity information (name, job type, qualification number) to ensure 100% power-on rate. The UWB positioning is used to capture the distance between the operator and the dangerous area of ​​the bridge erecting machine in real time. The dangerous area of ​​the bridge erecting machine includes the area under the outriggers and the area covered by the lifting equipment.

[0023] As an optional embodiment, the real-time sensing data uses GPS timing technology to synchronize the timestamps of all sensor data, synchronizing the sensor timestamps every 10 seconds. A time synchronization server is deployed on the local edge node, and a unified timestamp is added to all sensor data based on the NTP protocol to ensure that the time synchronization error of multi-source data is ≤1ms.

[0024] A local edge stage was deployed next to the bridge erecting machine's control room as an edge computing gateway. The edge computing gateway has an IP65 protection rating and is connected to the sensor acquisition module via a network cable. The gateway's network parameters were configured, and a connection was established with the project's 5G base station. Network latency (≤20ms) and packet loss rate (≤0.1%) were tested. Sensor timestamps were synchronized every 10 seconds via the gateway's GPS time synchronization function. A time synchronization server (NTP protocol) was deployed at the edge node to add a unified timestamp (format: YYYY-MM-DD HH:MM:SS.fff) to all sensor data, ensuring that the time synchronization error of multi-source data is ≤1ms.

[0025] Therefore, by configuring the gateway to cache locally, when the 5G network is interrupted, the data is automatically cached (caching time ≥ 24 hours), and automatically re-uploaded to the cloud platform after the network is restored.

[0026] As an optional embodiment, combined with Figure 2 As shown, in step 2, a 1:1 digital twin model is constructed based on the bridge erecting machine design drawings and BIM model. The collected real-time sensor data is mapped to the virtual model to achieve a visual presentation of the operation status, including: Step 2-1: Obtain the CAD design drawings, Revit BIM model, equipment rated load, structural dimensions, material performance parameters, and terrain data of the working scene for the bridge erecting machine; Step 2-2: Import the BIM model using Unity 3D software and recreate the detailed structure of the bridge erecting machine at a 1:1 scale, including the main beam, legs, lifting trolley, and lifting gear, with material parameters consistent with the actual structure. Steps 2-3: Build a work scenario model, including bridge piers, construction access roads, and surrounding environment; import terrain and elevation data to ensure that the scenario matches the actual working conditions. Steps 2-4: Real-time data mapping: Through the data interface API, real-time data transmission between edge nodes and the digital twin platform is realized, and the collected real-time sensor data is mapped to the corresponding positions of the model: stress data is presented as a color cloud map, blue ≤50MPa, green 50-100MPa, yellow 100-150MPa, red ≥150MPa, attitude data drives the model to move synchronously, personnel positions are marked with dynamic icons, and environmental data is displayed with numerical labels superimposed.

[0027] Furthermore, model verification can be performed to compare the state parameters of the virtual model with those of the actual equipment (such as the actual pitch angle of the spreader and the angle displayed by the model), with an error of ≤ ±0.2°; for simulating structural stress, the error between the stress calculated by the model and the actual sensor measurement value should be ≤ ±5%, and if they do not match, the model parameters should be adjusted.

[0028] By building a digital twin model of the entire process, the operation status can be visualized and traced, allowing managers to intuitively grasp the real-time status of equipment, personnel, and environment, providing intuitive support for risk assessment.

[0029] As an optional embodiment, combined with Figure 3 As shown, in step 3, a random forest algorithm is used to train a risk identification model using collected historical operational failure data. Based on this model, operational risks are identified from the real-time collected data, including: Step 3-1: Collect historical data of bridge erecting machine operation, including normal operation data and fault data. The fault data includes structural fatigue, attitude instability and personnel violation fault types. Divide the historical dataset into training set, validation set and test set in a ratio of 7:2:1. Step 3-2: Select 8 key features: maximum stress of the main beam, outrigger stress, pitch angle of the lifting device, roll angle, wind speed, visibility, distance between personnel and the danger zone, and operating load. Use the Pearson correlation coefficient method to remove redundant features (retain 1 feature with a correlation coefficient > 0.8); then standardize the selected features based on the Z-score formula to eliminate the influence of dimensions. Step 3-3: Build a random forest model based on the Scikit-learn library. Initial parameter settings: number of decision trees 100, maximum tree depth 10, minimum number of sample splits 2. Then, the model is trained using the aforementioned training set. The accuracy is tested with the validation set every 10 iterations. The parameters are tuned using the grid search method. The number of decision trees is 50-200 and the maximum tree depth is 5-15. Training is stopped when the accuracy of the validation set is ≥95%. The model was validated using the aforementioned test set, with an accuracy of ≥92% and a recall of ≥90%. If the accuracy was not met, the model parameters were readjusted and the model was retrained until the requirements were met. Steps 3-4: Deploy the trained risk identification model on local edge nodes to identify operational risks from real-time collected data and output operational risk levels of 0-3, where level 0 is no risk, level 1 is low risk, level 2 is high risk, and level 3 is extremely high risk.

[0030] As an optional embodiment, step 4, which involves constructing early warning systems for structural stress, attitude instability, and personnel safety distance boundaries, includes: Step 4-1: Construct a structural stress boundary early warning system. σ 预警 =[σ 允许 ] / k; Where, σ 允许The allowable stress of the main body material of the bridge erecting machine is 235MPa for Q345 steel and 170MPa for Q235 steel; k is the safety factor, k=2.0 for the main beam area and k=1.8 for the outrigger area; When the real-time stress σ 实时 ≥σ 预警 When σ triggers a Level 2 warning; 实时 ≥1.2×σ 预警 At that time, a Level 3 warning was triggered; Step 4-2: Construct an early warning system for attitude instability boundaries: θ 预警 =θ 额定 ×0.8; Where, θ 额定 The bridge erecting machine is designed with a maximum allowable pitch or roll angle; when the real-time attitude angle θ 实时 ≥θ 预警 When θ is triggered, a Level 2 warning is activated; when θ is triggered... 实时 ≥θ 额定 When the value reaches 0.95, a Level 3 warning is triggered.

[0031] Step 4-2: Establish personnel safety distance boundary early warning: The radius of the danger zone is set to R=5m. When the distance between personnel and the boundary of the danger zone is d≤1m, a level 2 warning is triggered; when d≤0.5m, a level 3 warning is triggered.

[0032] For example, consult the instruction manual of the bridge erecting machine to confirm the main material (Q345 steel / Q235 steel) and the design rated parameters (maximum pitch angle 15°, roll angle 8°, rated load P); refer to the "Safety Operation Procedures for Highway Bridge Erection Machines" (JTG / TF50-2011) to clarify the safety factor standard (main beam k=2.0, outrigger k=1.8).

[0033] As an optional real-time method, in step 5, based on the risk level output by the boundary warning and operational risk identification, a predetermined bridge erecting machine operation warning and safety control are executed, including: During a Level 1 warning, the operator will be alerted by an audible and visual alarm in the bridge erecting machine control room, such as a sound level of 80-90 dB and a red light flashing frequency of 2 Hz. During a Level 2 warning, the operator is alerted by an audible and visual alarm in the bridge erecting machine control room, and a notification is simultaneously sent to the safety officer's mobile app, prompting them to immediately complete a risk assessment. When a Level 3 warning is issued, the operator is alerted by an audible and visual alarm in the bridge erecting machine control room, and the alarm is simultaneously pushed to the safety officer's mobile APP. At the same time, the emergency shutdown procedure of the bridge erecting machine is automatically triggered, cutting off the lifting and traveling power circuits, activating the braking system, and sending a rescue notification to the project management platform, including real-time on-site images, personnel location, equipment status and other key on-site information.

[0034] As an optional embodiment, the method further includes the following steps: The operation path planning is optimized based on the digital twin model. When the ambient wind speed is ≥15m / s or the visibility is ≤300m, the model automatically recommends suspending the operation.

[0035] As an optional embodiment, the method further includes the following steps: By using personnel location data, real-time alerts can be issued to individuals who illegally approach dangerous areas.

[0036] As an optional embodiment, the method further includes the following steps: The digital twin model monitors equipment operating parameters in real time, sets warning thresholds for motor temperature and hydraulic pressure, and triggers equipment fault warnings when the real-time monitored sensor data exceeds the thresholds, reminding maintenance personnel to carry out timely repairs.

[0037] While the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the invention. Those skilled in the art can make various modifications and refinements without departing from the spirit and scope of the invention. Therefore, the scope of protection of the present invention shall be determined by the claims.

Claims

1. A method for monitoring the safety risks of a bridge erecting machine's operational status, characterized in that, Includes the following steps: Step 1: Based on the safety perception of the bridge erecting machine, collect real-time sensing data from the structural stress sensor, working posture sensor, working environment sensor and personnel positioning sensor deployed on the bridge erecting machine, and upload them to the local edge node and cloud platform. Step 2: Based on the bridge erecting machine design drawings and BIM model, construct a 1:1 digital twin model, and map the collected real-time sensor data to the virtual model to realize the visualization of the operation status; Step 3: Use the random forest algorithm to train the risk identification model using historical operation failure data, and use the risk identification model to identify operation risks based on the real-time collected data, outputting operation risk levels 0-3, where level 0 is no risk, level 1 is low risk, level 2 is high risk, and level 3 is extremely high risk. Step 4: Construct early warning systems for structural stress, attitude instability, and personnel safety distance boundaries. Perform risk identification on the real-time collected data and output risk levels from 0 to 3, where level 0 is no risk, level 1 is low risk, level 2 is high risk, and level 3 is extremely high risk. Step 5: Based on the risk level output from boundary warnings and operational risk identification, execute the predetermined bridge erecting machine operation warnings and safety controls; In step 2, a 1:1 digital twin model is constructed based on the bridge erecting machine design drawings and BIM model. The collected real-time sensor data is mapped to the virtual model to achieve a visual presentation of the operation status, including: Step 2-1: Obtain the CAD design drawings, Revit BIM model, equipment rated load, structural dimensions, material performance parameters, and terrain data of the working scene for the bridge erecting machine; Step 2-2: Import the BIM model using Unity 3D software and recreate the detailed structure of the bridge erecting machine at a 1:1 scale, including the main beam, legs, lifting trolley, and lifting gear, with material parameters consistent with the actual structure. Steps 2-3: Build a work scenario model, including bridge piers, construction access roads, and surrounding environment; import terrain and elevation data to ensure that the scenario matches the actual working conditions. Steps 2-4: Real-time data mapping: Through the data interface API, real-time data transmission between edge nodes and the digital twin platform is realized, and the collected real-time sensor data is mapped to the corresponding positions of the model: stress data is presented as a color cloud map, blue ≤50MPa, green 50-100MPa, yellow 100-150MPa, red ≥150MPa, attitude data drives the synchronous movement of the model, personnel positions are marked with dynamic icons, and environmental data is displayed with numerical labels superimposed. Step 4 involves establishing early warning systems for structural stress, attitude instability, and personnel safety distance boundaries, including: Step 4-1: Construct a structural stress boundary early warning system. s 预警 =[σ 允许 ] / k; Where, σ 允许 The allowable stress of the main body material of the bridge erecting machine is 235MPa for Q345 steel and 170MPa for Q235 steel; k is the safety factor, k=2.0 for the main beam area and k=1.8 for the outrigger area; When the real-time stress σ 实时 ≥σ 预警 When σ triggers a Level 2 warning; 实时 ≥1.2×σ 预警 At that time, a Level 3 warning was triggered; Step 4-2: Construct an early warning system for attitude instability boundaries: i 预警 =θ 额定 ×0.8; Where, θ 额定 The bridge erecting machine is designed with a maximum allowable pitch or roll angle; when the real-time attitude angle θ 实时 ≥θ 预警 When θ is triggered, a Level 2 warning is activated; when θ is triggered... 实时 ≥θ 额定 When the value is ×0.95, a Level 3 warning is triggered; Step 4-2: Establish personnel safety distance boundary early warning: The radius of the danger zone is set to R=5m. When the distance between personnel and the boundary of the danger zone is d≤1m, a level 2 warning is triggered; when d≤0.5m, a level 3 warning is triggered.

2. The method for monitoring the safety risks of bridge erecting machines during operation according to claim 1, characterized in that, In step 1, the structural stress sensor is a fiber optic grating type sensor, which is installed at the mid-span of the main beam and at the connection of the support legs, with a sampling frequency of 10Hz. The working attitude sensor adopts an IMU inertial measurement unit, which is installed on the crane trolley and the lifting device to monitor the pitch angle and roll angle, with a sampling frequency of 10Hz; The environmental sensor is deployed at a high point in the work area to monitor wind speed, visibility, and temperature, with a sampling frequency of 10Hz; The personnel positioning sensor uses UWB positioning and equips the operators with positioning tags to capture the distance between the personnel and the dangerous area of ​​the bridge erecting machine in real time. The dangerous area of ​​the bridge erecting machine includes the area under the outriggers and the area covered by the spreader.

3. The method for monitoring the operational status safety risks of a bridge erecting machine according to claim 1, characterized in that, The real-time sensing data uses GPS timing technology to synchronize the timestamps of all sensor data. The sensor timestamps are synchronized every 10 seconds. A time synchronization server is deployed on the local edge node, and a unified timestamp is added to all sensor data based on the NTP protocol to ensure that the time synchronization error of multi-source data is ≤1ms.

4. The method for monitoring the safety risks of bridge erecting machines during operation according to claim 1, characterized in that, In step 3, a random forest algorithm is used to train a risk identification model using collected historical operational failure data. Based on this model, operational risks are identified from the real-time collected data, including: Step 3-1: Collect historical data of bridge erecting machine operation, including normal operation data and fault data. The fault data includes structural fatigue, attitude instability and personnel violation fault types. Divide the historical dataset into training set, validation set and test set in a ratio of 7:2:

1. Step 3-2: Select 8 key features: maximum stress of main beam, outrigger stress, pitch angle of lifting equipment, roll angle, wind speed, visibility, distance between personnel and danger zone, and operating load. Use the Pearson correlation coefficient method to eliminate redundant features; then standardize the selected features based on the Z-score formula. Step 3-3: Build a random forest model based on the Scikit-learn library. Initial parameter settings: number of decision trees 100, maximum tree depth 10, minimum number of sample splits 2. Then, the model is trained using the aforementioned training set. The accuracy is tested with the validation set every 10 iterations. The parameters are tuned using the grid search method. Training is stopped when the accuracy on the validation set is ≥95%. The model was validated using the aforementioned test set, with an accuracy of ≥92% and a recall of ≥90%. If the accuracy was not met, the model parameters were readjusted and the model was retrained until the requirements were met. Steps 3-4: Deploy the trained risk identification model on local edge nodes to identify operational risks from real-time collected data and output operational risk levels of 0-3, where level 0 is no risk, level 1 is low risk, level 2 is high risk, and level 3 is extremely high risk.

5. The method for monitoring the operational status safety risks of a bridge erecting machine according to claim 1, characterized in that, In step 5, based on the risk level output from boundary warning and operational risk identification, predetermined bridge erecting machine operation warning and safety control are implemented, including: During a Level 1 warning, the operator will be alerted via an audible and visual alarm in the bridge erecting machine's control room. During a Level 2 warning, the operator is alerted by an audible and visual alarm in the bridge erecting machine control room, and a notification is simultaneously sent to the safety officer's mobile app, prompting them to immediately complete a risk assessment. When a Level 3 warning is issued, the operator is alerted by an audible and visual alarm in the bridge erecting machine's control room. The alarm is also simultaneously pushed to the safety officer's mobile app and automatically triggered to shut down the bridge erecting machine's emergency stop procedure, cutting off the lifting and traveling power circuits and sending a rescue notification to the project management platform.

6. The method for monitoring the operational status safety risks of a bridge erecting machine according to claim 1, characterized in that, The method further includes the following steps: The operation path planning is optimized based on the digital twin model. When the ambient wind speed is ≥15m / s or the visibility is ≤300m, the model automatically recommends suspending the operation.

7. The method for monitoring the safety risks of bridge erecting machines during operation according to claim 1, characterized in that, The method further includes the following steps: By using personnel location data, real-time alerts can be issued to individuals who illegally approach dangerous areas.

8. The method for monitoring the safety risks of bridge erecting machines during operation according to claim 1, characterized in that, The method further includes the following steps: The digital twin model monitors equipment operating parameters in real time, sets warning thresholds for motor temperature and hydraulic pressure, and triggers equipment fault warnings when the real-time monitored sensor data exceeds the thresholds, reminding maintenance personnel to carry out timely repairs.

Images