Early warning method and system for unsafe lifting behavior of cranes based on digital twins

Abstract

The present invention belongs to the field of hoisting operation safety control, and specifically discloses a method and system for early warning of unsafe hoisting behavior of cranes based on digital twins, which includes a model construction stage: constructing a digital twin model according to the crane BIM model and environmental attribute data; establishing an early warning module, and embedding the early warning module into the digital twin model; the early warning module includes multiple graph structures, each of which corresponds to different unsafe hoisting behaviors; each node in the graph structure represents a hoisting parameter, and the state of the node represents a hoisting action state; the nodes, the states of the nodes, and the connection relationships between the nodes are combined to form a graph structure. Model application stage: the hoisting parameters are obtained in real time through sensors and mapped to the model, so as to judge the state of the nodes, and then determine whether unsafe hoisting behavior occurs, so as to realize early warning of unsafe hoisting behavior. The present invention can timely warn of potential unsafe hoisting behaviors, and has the advantages of real-time, accuracy and high efficiency.

Description

Technical Field

[0001] The present invention belongs to the field of safety control of hoisting operations, and more specifically, relates to a method and system for early warning of unsafe hoisting behavior of cranes based on digital twins. Background Art

[0002] Tower cranes are widely used in the construction field. However, due to the complexity of the lifting project itself, the differences in operator skills and experience, and environmental factors, the safety risks are difficult to avoid. Once an accident occurs, it will not only threaten personal safety, but also cause huge losses to property. Therefore, it is very important to prevent unsafe lifting behaviors of tower cranes.

[0003] At present, the research on hoisting operation safety management mainly focuses on the interaction between the crane and the surrounding environment, the crane itself, etc. Based on sensors, machine learning and other technologies are used to build early warning models. However, there are still few studies on the operating behavior habits of tower cranes. Therefore, it is very necessary to provide a new method and system for early warning of unsafe hoisting behavior of tower cranes. Summary of the invention

[0004] In response to the above defects or improvement needs of the prior art, the present invention provides a method and system for early warning of unsafe lifting behaviors of cranes based on digital twins, which aims to achieve timely and accurate early warning of various unsafe lifting behaviors of cranes.

[0005] To achieve the above-mentioned purpose, according to a first aspect of the present invention, a method for early warning of unsafe hoisting behavior of a crane based on digital twin is proposed, which includes a model building stage and a model application stage, wherein:

[0006] Model building phase:

[0007] Obtain crane BIM model and environmental attribute data, and install sensors for obtaining lifting parameters;

[0008] Build a digital twin model based on the crane BIM model and environmental attribute data;

[0009] Establish an early warning module based on a graph structure, and embed the early warning module into the digital twin model, so that the sensor data interacts with the digital twin model through the nodes in the graph structure; the early warning module includes multiple graph structures, each graph structure corresponds to a different unsafe lifting behavior; each node in the graph structure represents a lifting parameter, and the state of the node represents a lifting action state; according to the operation process of the unsafe lifting behavior, select relevant nodes, and combine the nodes, the states of the nodes and the connection relationship between the nodes to form a graph structure;

[0010] Model application phase:

[0011] During the crane hoisting process, the hoisting parameters are obtained in real time through sensors. The early warning module judges the state of the node based on the hoisting parameters, and then determines whether any unsafe hoisting behavior occurs, and displays it in the digital twin model to achieve early warning of unsafe hoisting behavior of the crane.

[0012] As a further preference, the hoisting parameters include: weight of hoisted object, hoisting height, hoisting speed, rotation angle, rotation speed, amplitude change distance, amplitude change speed, and swing angle.

[0013] As a further preference, the environmental attribute data includes terrain information of the crane construction site, which is acquired by scanning with a laser radar mounted on an unmanned aerial vehicle.

[0014] As a further preferred embodiment, the lifting parameters corresponding to each node are obtained through sensors, and the sensor data are deduplicated, standardized and processed for outliers.

[0015] As a further preference, the connection relationship between the nodes represents the relationship between the lifting actions, which includes the spatial position relationship, the temporal sequence and the process sequence of the actual operation.

[0016] As a further preferred embodiment, each unsafe lifting behavior is pre-classified, so that in the model application stage, a graded warning is issued according to the unsafe lifting behavior determined to have occurred.

[0017] According to a second aspect of the present invention, a digital twin-based crane unsafe lifting behavior warning system is provided, which includes a processor, and the processor is used to execute the above-mentioned digital twin-based crane unsafe lifting behavior warning method.

[0018] According to a third aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned digital twin-based crane unsafe lifting behavior warning method is implemented.

[0019] In general, the above technical solution conceived by the present invention has the following technical advantages compared with the prior art:

[0020] 1. The present invention utilizes real-time sensor data to map relevant data into a virtual model, realizing virtual-real synchronous mapping between the digital twin virtual model and the physical entity, thereby improving data transmission efficiency and timeliness; and integrating an unsafe lifting behavior warning module based on a graph structure into the digital twin model, enhancing the function of the digital twin model, realizing refined monitoring and early warning of lifting behavior, improving the accuracy and timeliness of early warning, effectively reducing the risk of lifting operations, and ensuring personnel safety.

[0021] 2. The early warning system of the present invention can automatically screen hoisting events and send different levels of early warning information to relevant responsible parties according to the early warning level, so that the early warning delivery is more accurate and invalid early warnings are avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of a method for early warning of unsafe hoisting behavior of a crane based on digital twins according to an embodiment of the present invention;

[0023] Figure 2 Schematic diagram of a graph structure of a common unsafe lifting event of a tower crane according to an embodiment of the present invention, wherein (a) to (g) represent graph structures corresponding to lifting events 1 to 7, respectively. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0025] An embodiment of the present invention provides a method for early warning of unsafe hoisting behavior of a crane based on digital twins, such as Figure 1 As shown, the following steps are included:

[0026] S1. Data collection and processing: Data related to the physical entity of the tower crane is collected through on-site surveys, on-site sensors, monitoring systems, etc., mainly including:

[0027] Obtain topographic information of the construction site, i.e. environmental attribute data, through drones equipped with laser radar and other technologies;

[0028] Various hardware equipment is deployed at appropriate locations such as tower cranes and construction sites to meet the monitoring and data collection needs of hoisting operations, including but not limited to slewing sensors, amplitude sensors, height sensors, weighing sensors, Beidou GNSS, inertial sensors, wind speed sensors, wind direction sensors, etc.;

[0029] Furthermore, after data collection is completed, the data is preprocessed and cleaned, including data deduplication, data standardization, outlier processing, etc., to ensure the accuracy and consistency of the data, improve the quality and credibility of the data, and provide a reliable foundation for subsequent data analysis and modeling.

[0030] S2. Construction of digital twin model, including the construction of digital twin virtual model and the realization of virtual-reality synchronous mapping: In the digital twin platform, the digital twin virtual model is constructed by integrating environmental attribute data, BIM model and other information; a complete set of mapping rules is established to associate real-time sensor data with the digital twin virtual model to drive the movement of the digital twin virtual model.

[0031] The specific steps for building a digital twin model are as follows:

[0032] The first step is to establish a virtual model: In UE4, use laser scanning point cloud to import information such as the terrain of the construction site, establish an environmental scene, and restore the construction site environment; determine the tower crane model based on construction organization design and other documents, and use SolidWorks and other software to establish a BIM model of the tower crane to accurately represent the structure and motion characteristics of the tower crane.

[0033] The second step is base point calibration and import: calibrate the base point of the tower crane operation in the environmental scene, determine its position, import the BIM model of the tower crane, and adjust the imported BIM model by using transformation tools (such as move, rotate, and scale) to ensure that the position, orientation, and size of the model are consistent with the actual crane in the environmental scene.

[0034] The third step is to establish wireless communication: establish a wireless communication architecture based on the MQTT protocol. The STM32 microcontroller controls the ESP8266 module to connect to the wireless access point (AP), and then connects to the MQTT server to realize data uplink and send sensor data to the server. UE4 obtains the uplink sensor data by grabbing data from the server.

[0035] Step 4: Establish mapping rules: In the UE4 blueprint, write logic by adding events, connecting nodes, and setting function properties to associate the received sensor data with the various components of the digital twin virtual model, drive the digital twin virtual model to move, and keep it synchronized with the state of the tower crane. For each function, configure its properties as needed. For example, for data processing functions, set parsing rules, data formats, and conversion options to adapt them to the properties of the digital twin model. For example, IMU sensor data needs to achieve real-time synchronization between the model and sensor data by setting the model's position, rotation, and scaling properties. Laser ranging sensor data needs to be proportionally changed to ensure that the data is correctly associated with the model properties. In the data mapping process, consider possible abnormal situations and errors, add appropriate error handling and fault tolerance mechanisms to ensure the stability and reliability of the system. For example, when processing data with abnormal values, the abnormal data can be eliminated, and when data loss occurs, a reasonable interpolation method can be used for processing.

[0036] S3, establish an early warning module based on a graph structure and embed it into the digital twin model described in S2;

[0037] To establish an early warning module based on a graph structure, the specific steps are as follows:

[0038] The first step is to construct nodes. As shown in Table 1, each node represents different lifting parameters or attributes, such as hoisting weight (HW), hoisting height (HH), hoisting speed (HS), rotation angle (RA), rotation speed (RS), amplitude change distance (LD), amplitude change speed (LS), swing angle (SA), etc. Different from the traditional graph structure algorithm, the node definition of the present invention is more flexible. The node represents the sensor data related to the node at the data level, and has a link relationship with the sensor data on the corresponding component. The sensor data interacts with the digital twin model through the node; at the physical level, the node represents the various parameters and corresponding components in the lifting operation, such as lifting, rotation, amplitude change and other operations.

[0039] The second step is to establish connections between nodes. This process requires comprehensive consideration of the spatial position relationship, temporal order, and actual operation process. Node connection is a multi-level relationship used to establish the association relationship between nodes in the graph structure, which represents the interaction, dependency, and connection relationship between nodes. It includes not only the spatial position relationship between nodes, but also the temporal order (the operation of one node occurs before or after another node), and the actual operation process (the operation order and interaction of each node during the lifting process); node connection represents the relationship between the collected data of each node over time at the data level, and represents the order of operation of each node during the lifting process at the physical level.

[0040] The third step is to determine the status of each node according to the node attributes and the set judgment rules, including whether the lifting height is within the warning boundary, whether there is an abnormality in the rotation angle, whether the amplitude change distance exceeds the limit, etc., as shown in Table 1.

[0041] Table 1 Node attributes of the unsafe lifting behavior graph based on graph structure

[0042]

[0043] The fourth step is to combine the judgment logic of different lifting events, according to the timeline and operation process of the lifting event, combine each node, node state and the connection relationship between nodes to form a complete graph structure to identify different lifting states and behaviors. At the data level, the graph structure represents different data relationships and data flows, and the connection relationship between nodes represents the interaction and transmission between different data; at the physical level, the graph structure represents different lifting behaviors, each node and node state corresponds to a specific lifting action and state in the lifting process, and the connection relationship between nodes represents the sequence and dependency between lifting behaviors. Through the construction of this graph structure, the order and evolution of each action in the lifting process can be accurately described.

[0044] Taking the common unsafe tower crane hoisting events shown in Table 2 as an example, according to relevant safety regulations and laws, the causal factors and processes of these events are analyzed, the judgment logic of different hoisting events is sorted out, and according to the timeline and operation process of the hoisting event, the relevant nodes are selected, and the nodes, node states and connection relationships between nodes are combined to construct a Figure 2 The graph structure shown is used to identify different lifting states and behaviors, where Times represents the time when each node and node state of the lifting event is triggered.

[0045] Table 2 Common unsafe hoisting behaviors of tower cranes

[0046] Serial number Unsafe lifting behavior Hoisting event 1 Suspended objects are suspended in the air for a long time Lifting incident 2 There is no obvious deceleration process when the heavy object is in place / approaching the limit Hoisting event 3 No test lift when the load exceeds 95% Lifting event 4 Working close to the alarm value for a long time Hoisting event 5 Brutal Operation Hoisting event 6 Reverse rotation Hoisting event 7 The hanging object swings obviously

[0047] For ease of understanding, the following uses hoisting events 1 and 2 as examples to illustrate the graph structure and the determination of unsafe hoisting behaviors:

[0048] Lifting event 1 represents that the suspended object is suspended in the air for a long time. After sorting out the lifting event 1, its occurrence involves the HW, HH, HS, RS, and LS nodes. The spatial position relationship, temporal sequence, and actual operation process of these nodes are comprehensively considered to establish the connection between the nodes; at the same time, the status of each node is determined, and combined with the judgment logic of lifting event 1, according to the timeline and operation process of the lifting event, each node, node status, and the connection relationship between nodes are combined to form a graph structure to represent lifting event 1. When HW maintains in the S1 state for more than the specified time, it means that there is a suspended object hanging on the hook; when HH maintains in the S1 state for more than the specified time, it means that the hook is stagnant in the air; when HS, RS, and LS maintain in the S1 state for more than the specified time, it means that no lifting, rotation, amplitude change, etc. operations have been performed; based on the above results, it can be determined that lifting event 1 has occurred, that is, the suspended object is suspended in the air for a long time.

[0049] Lifting event 2 means that there is no obvious deceleration process when the heavy object is in place / approaching the limit. Lifting event 2 is sorted out, and its occurrence involves HH, HS, RA, RS, LD, and LS nodes. The spatial position relationship, temporal sequence, and actual operation process of these nodes are comprehensively considered to establish the connection between the nodes; at the same time, the status of each node is judged, combined with the judgment logic of lifting event 2, according to the timeline and operation process of the lifting event, each node, node status, and the connection relationship between nodes are combined to form a graph structure to represent lifting event 2. When HH maintains the S3 state for more than the specified time, HS is judged. If HS also maintains the S3 state for more than the specified time, it means that the hook is still in a high-speed state when approaching the limit; when RA maintains the S3 state for more than the specified time, RS is judged. If RS also maintains the S3 state for more than the specified time, it means that the slewing mechanism is still in a high-speed state when approaching the limit; when LD maintains the S3 state for more than the specified time, LS is judged. If LS also maintains the S3 state for more than the specified time, it means that the boom mechanism is still in a high-speed state when approaching the limit; Based on the above results, it can be determined that lifting event 2 occurs, that is, there is no obvious deceleration process when the heavy object is in place / approaching the limit.

[0050] S4. Early warning process and information management: collect key data in the lifting process through sensors, input the data into the digital twin model and early warning module for analysis and judgment, and output corresponding early warning information according to the preset early warning rules; when unsafe lifting behavior occurs, the lifting events can be automatically screened, and different levels of early warning information can be sent to relevant responsible parties according to the early warning level (such as blue / yellow / orange / red).

[0051] Specifically, when the early warning module determines that unsafe lifting behavior has occurred based on relevant sensor data, the early warning system will send information from the mobile device to operators, managers and other relevant responsible parties, informing them of the unsafe lifting behavior and the events and locations where it occurred, urging them to correct it as soon as possible, and record the lifting events, early warning information and subsequent feedback for analysis and processing; at the same time, the early warning system will compile statistics on daily early warning information and form statistical reports so that the relevant responsible parties can understand and grasp the lifting situation and take timely measures to rectify and improve it.

[0052] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for early warning of unsafe crane lifting behavior based on digital twin, characterized in that: It includes the model building phase and the model application phase, in which: Model building phase: Obtain crane BIM model and environmental attribute data, and install sensors for obtaining lifting parameters; Build a digital twin model based on the crane BIM model and environmental attribute data; Establish an early warning module based on a graph structure, and embed the early warning module into the digital twin model, so that the sensor data interacts with the digital twin model through the nodes in the graph structure; the early warning module includes multiple graph structures, each graph structure corresponds to a different unsafe lifting behavior; each node in the graph structure represents a lifting parameter, and the state of the node represents a lifting action state; according to the operation process of the unsafe lifting behavior, select relevant nodes, and combine the nodes, the states of the nodes and the connection relationship between the nodes to form a graph structure; Model application phase: During the crane hoisting process, the hoisting parameters are obtained in real time through sensors. The early warning module judges the state of the node based on the hoisting parameters, and then determines whether any unsafe hoisting behavior occurs, and displays it in the digital twin model to achieve early warning of unsafe hoisting behavior of the crane.

2. The method for early warning of unsafe crane hoisting behavior based on digital twins according to claim 1 is characterized in that: The hoisting parameters include: hoisting weight, hoisting height, hoisting speed, rotation angle, rotation speed, amplitude change distance, amplitude change speed, and swing angle.

3. The method for early warning of unsafe crane hoisting behavior based on digital twins according to claim 1, characterized in that: The environmental attribute data includes terrain information of the crane construction site, which is obtained by scanning with a laser radar mounted on an unmanned aerial vehicle.

4. The method for early warning of unsafe crane hoisting behavior based on digital twins according to claim 1, characterized in that: The corresponding lifting parameters of each node are obtained through sensors, and the sensor data are deduplicated, standardized and processed for outliers.

5. The method for early warning of unsafe crane hoisting behavior based on digital twins according to claim 1, characterized in that: The connection relationship between the nodes represents the relationship between the lifting actions, which includes the spatial position relationship, the temporal sequence and the process sequence of the actual operation.

6. The method for early warning of unsafe crane hoisting behavior based on digital twins according to any one of claims 1 to 5, characterized in that: Each unsafe lifting behavior is pre-classified, so that during the model application stage, graded warnings can be issued based on the unsafe lifting behaviors that have occurred.

7. A crane unsafe lifting behavior early warning system based on digital twin, characterized in that: It includes a processor, which is used to execute the digital twin-based crane unsafe lifting behavior early warning method as described in any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the method for early warning of unsafe lifting behavior of a crane based on digital twin as described in any one of claims 1 to 6.

Images