A construction site hoisting anti-collision method based on digital twin

Abstract

The present invention belongs to the field of anti-collision systems, and specifically discloses a construction site lifting and hoisting anti-collision method based on digital twins, the method comprising the following steps: S1, analysis of collision problems of multiple types of machines; S2, establishment of anti-collision models and thresholds; S3, establishment of digital twin models; S4, interaction between models and the site; S5, continuity judgment; S6, determination of alarm types; S7, response strategies; This application combines digital twin technology and Beidou satellite RTK technology fusion algorithms to construct a multi-machine flow, three-dimensional, real-time anti-collision system that can realize multi-party real-time supervision, anti-collision, real-time alarm, and data storage and recording functions during the operation of multiple types of lifting equipment, thereby achieving real-time consistency between information flow and business flow.

Description

Technical Field

[0001] The present invention relates to the field of anti-collision systems, and in particular to a construction site hoisting anti-collision method based on digital twins. Background Art

[0002] With the development of modern architectural design and construction technology, there are more and more large buildings such as large exhibition halls, bridges, nuclear power plants and various large group buildings. Correspondingly, the machinery and equipment in the construction industry are also increasingly developing towards large-scale and automation. With the gradual application of crawler-type multi-machine equipment in linear engineering and nuclear power fields, the safety management of tower cranes and crawler-type multi-machines is also facing more severe tests. Compared with ordinary construction projects, in the application scenarios of nuclear power projects, there will be more different types of tower cranes and other construction machinery in limited construction areas. The dense arrangement of tower cranes will inevitably lead to a sharp increase in the risk of cross-collision of multi-machine operations, which puts higher requirements on the anti-collision safety management of tower crane group operations.

[0003] In the early days, the anti-collision research between tower cranes, truck cranes, crawler cranes and other types of machines was mainly carried out through the research of mechanical limit devices of tower cranes, including weight limiters, torque limiters, lifting limiters, etc. When the limiter exceeds the threshold, an alarm is triggered, which indirectly realizes the anti-collision monitoring function of the tower crane operation process. However, due to its single structure and function, it is difficult to effectively ensure the safety of multiple types of operations and cannot meet the increasing safety monitoring needs of users. SMIE's AC30 system dynamic position calculation and risk estimation technology, POTAIN's tower crane area restriction planning technology, e-build's TAC-3000 tower crane group anti-collision work boundary restriction and area protection technology are all mature technical representatives of tower crane anti-collision. However, these tower crane safety monitoring technologies are all passive defense safety models, and there is still room for improvement in initiative, intelligence, real-time and rapidity.

[0004] Therefore, studying the monitoring methods to ensure the collision avoidance of multiple types of aircraft is of great significance to promoting the operation management of multiple types of aircraft. Summary of the invention

[0005] The purpose of the present invention is to solve the defects of the prior art and provide a construction site lifting and hoisting anti-collision method based on digital twin.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0007] A construction site hoisting anti-collision method based on digital twins includes the following steps:

[0008] S1. Analysis of collision problems of multiple types of machines: During the construction process, the main factors and data parameters affecting the occurrence of collisions are determined by consulting literature and analyzing specific cases;

[0009] S2. Establishing anti-collision model and threshold: Based on the main collision scenarios and data parameters affecting collision obtained in S1, a set of anti-collision models covering all scenarios is established, and thresholds are set;

[0010] S3. Digital twin model establishment: Collect key data based on the current technical background and the problem to be studied to map it into a virtual model;

[0011] S4, Model and Site Interaction: After the digital twin data model of multiple models is established, the data of physical entities are continuously collected to achieve model calibration and timely status update;

[0012] S5. Continuity judgment: judge the error between the physical entity and the digital twin model. When the error is greater than the threshold, the state of the model is considered discontinuous, and further data needs to be extracted to train the model to integrate key influencing features and associations;

[0013] When the error is less than the threshold, the anti-collision model established based on the real-time data collected by the physical device and the current state of the virtual model predicts the possibility of a collision;

[0014] S6. Determine the alarm type: The model judged by continuity is considered to have high accuracy and real-time performance. When there is a risk of collision between multiple types of machines, the collision type and threshold established by the anti-collision model are used to determine the collision type and issue different types of alarms.

[0015] S7. Response strategy: According to the requirements of the operation specifications of multiple machine types, when different types of alarms occur, the project site adopts timely and effective response strategies to eliminate the alarm information.

[0016] Furthermore, in S2, building an anti-collision model includes the following steps:

[0017] S21, Multi-machine data collection:

[0018] S22. Establishing a collision avoidance model at the same level:

[0019] S23. Establish a height difference anti-collision model.

[0020] Further, S21, data collection of multiple machine types, including: hoisting visualization through the hoisting area camera on the hoisting equipment, hoisting equipment arm length data collection through the boom length sensor on the hoisting equipment, hoisting equipment vertical rotation angle data collection through the boom angle sensor on the hoisting equipment, positioning the on-site position of each machine type and measuring the boom rotation angle through the Beidou positioning module on the hoisting equipment, hook rope quality monitoring through the winch mechanism camera on the hoisting equipment, hook length detection through the hook height sensor on the hoisting equipment, and data processing and calculation through the host on the hoisting equipment; wherein, the hoisting equipment includes tower cranes and mobile lifting machinery.

[0021] Further, S22, a collision avoidance model at the same level is established, and the method used is:

[0022] Select a point O on the ground of the construction site as the origin to establish the global reference coordinate system {O-xyz}, with the east as the x-axis, the north as the y-axis, and the vertical direction as the z-axis, and derive the position matrix of the lifting machinery;

[0023] Assume that two adjacent tower cranes TJ1 and TJ2 are arranged at the construction site, and the booms are operating at the same level. In this case, the boom has the longest radius, and its outer edge is the front end where the collision occurs;

[0024] Assuming that the boom lengths of TJ1 and TJ2 are R1 and R2 respectively, when O1O2>R1+R2, there is no risk of cross operation between the booms;

[0025] When O1O2≤R1+R2, there is an intersecting operation area T. When the two tower crane arms move to the intersecting area, there is a risk of collision. According to the set anti-collision threshold, an anti-collision warning is issued. This anti-collision model is also applicable when the tower crane and the mobile lifting machinery are performing hoisting operations at the same horizontal height.

[0026] Further, S23, a height difference anti-collision model is established. For two tower cranes, the method used is:

[0027] Assume that two adjacent tower cranes TJ1 and TJ2 are arranged at the construction site, operating at different levels. Assume that TJ1 is a low-level tower crane and TJ2 is a high-level tower crane;

[0028] There is no collision risk for each boom. Assume that the operating heights of TJ1 and TJ2 are H1 and H2 respectively, the hook lengths are L1 and L2 respectively, and the trolley amplitudes are r1 and r2 respectively;

[0029] When O1O2>R1+r2, there is no risk of collision between tower cranes; when O1O2≤R1+r2, there is an intersection area between the two tower cranes in the two-dimensional space when viewed from above, but when H2-L2>H1, there is a height difference in the cross operation, and there is also no risk of collision. When H2-L2≤H1, there is a cross operation area T, and there is a risk of collision when the hooks of the two tower cranes move to the intersection area. This anti-collision model is also applicable when tower cranes and mobile lifting machinery are performing hoisting operations at different levels.

[0030] Further, S23, a height difference anti-collision model is established. For tower cranes and truck cranes, the method used is:

[0031] Assume that the height of the mobile crane is H2, the arm length is R2, the horizontal rotation angle is θ, the tower crane is in a low position, and the mobile crane is in a high position. When O1O2>R1+R2cosθ, there is no collision risk.

[0032] When O1O2≤R1+R2cosθ, there is an intersection area between the tower crane and the truck crane in the two-dimensional space viewed from above, but when H2+R2sinθ-L2>H1, there is a height difference in the cross-operation, and there is also no collision risk. When H2+R2sinθ-L2≤H1, there is a cross-operation area T, and there is a collision risk when the tower crane and the truck crane hook operate to the intersection area.

[0033] Furthermore, the thresholds include anti-collision warning and alarm thresholds. Based on the operation model of various types of lifting equipment, combined with the braking process of the lifting equipment, the selection of warning and alarm thresholds is considered, without considering the interference of external conditions such as wind speed and load. Assuming that the tower crane or mobile lifting mechanical braking rotation angle is w, the braking distance is L, and when braking, the variable frequency speed control method is adopted. Assuming that there are n gears in total, the braking rotation angles under each gear are w and l, respectively. i , then w=∑w i , i=1...n, where the braking rotation angle in each gear is related to factors such as the state, model, and load of the crane type. It is calculated based on actual equipment parameters, and the anti-collision warning and alarm thresholds are set in the system.

[0034] Furthermore, when calculating the anti-collision threshold between the tower crane and the truck crane, the truck crane runs counterclockwise, and there is a risk of collision between the truck crane's lifting wire rope and the tower crane's lifting arm. Then L = 2R2sin(∑w i / 2), the alarm distance can be set in the system as L alarm ≥ L, the warning distance is magnified by a certain coefficient based on the actual situation on site, which can be set to For the amplification system.

[0035] The beneficial effects of the present invention are as follows: the present application combines digital twin technology and Beidou satellite RTK technology to construct a multi-machine mobile anti-collision system that can realize multi-party real-time supervision, anti-collision, real-time alarm, and data storage and recording functions during the operation of multi-machine lifting equipment, achieving real-time consistency between information flow and business flow. The main advantages are as follows:

[0036] (1) Realize on-site 3D visualization, greatly improving supervision efficiency; the equipment in the digital twin system corresponds to the location and action of on-site equipment one by one, meeting the needs of real-time positioning of mobile targets, further improving on-site supervision efficiency;

[0037] (2) Redundant verification and synchronous alarm effectively avoid blind spots in supervision. The anti-collision server implements redundant verification of pre-alarm information, and the server and the driver can give synchronous alarms to avoid blind spots in supervision.

[0038] (3) The digital twin solution is used to ensure the accuracy of the results. The crane corresponds 1:1 to the actual model in the digital twin system, and there is no need to enlarge the crane boom range, making data calculation more realistic. The large model of the anti-collision server and the small model of the data acquisition device perform synchronous calculations and mutual verification, further ensuring the accuracy of the results. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a schematic diagram of the anti-collision process based on digital twins of the present invention;

[0040] Figure 2 This is a simplified model diagram of the flat-arm tower crane of the present invention;

[0041] Figure 3 This is a schematic diagram of two tower cranes operating at the same level;

[0042] Figure 4 It is a three-dimensional schematic diagram of two tower cranes operating at different levels;

[0043] Figure 5 It is a two-dimensional side view of two tower cranes operating at different levels;

[0044] Figure 6 It is a two-dimensional side view of tower cranes and truck cranes operating at different levels;

[0045] Figure 7 A top view of the anti-collision alarm;

[0046] Figure 8 It is a mobile multi-machine collision avoidance system architecture based on digital twins. DETAILED DESCRIPTION

[0047] Reference Figure 1 A construction site hoisting anti-collision method based on digital twins, characterized in that it includes the following steps:

[0048] S1: Analysis of collision problems of multiple types of machines. During the construction process, there are many scenarios where collisions occur. By consulting literature and analyzing specific cases, we can determine the main factors and data parameters that affect the occurrence of collisions, and then deal with collision safety issues in a more targeted manner.

[0049] S2: Establishing anti-collision model and threshold. Based on the main collision scenarios and data parameters affecting collisions obtained in step 1, a set of anti-collision models covering all scenarios is established, and collision thresholds are set to achieve the purpose of anti-collision warning.

[0050] S3: Establishment of digital twin model. The digital twin model of multiple models does not need to collect all the detailed data of the physical entity to achieve complete virtual mapping, but collects as much key data as possible in the current technical context and combines the problems to be studied to map the virtual model, which meets the research needs while being timely and accurate.

[0051] S4: Model-site interaction. After the digital twin data model of multiple models is established, the data of physical entities is continuously collected to realize model calibration and timely status update, thereby achieving the goal of high accuracy of the digital twin model.

[0052] S5: Continuity judgment. Judge the error between the physical entity and the digital twin model. When the error is greater than the threshold, the state of the model is considered discontinuous, and further data needs to be extracted to train the model to integrate key influencing features and associations. When the error is less than the threshold, the anti-collision model established based on the real-time data collected by the physical device and the current state of the virtual model predicts the possibility of a collision.

[0053] S6: Determine the alarm type. The model that passes the continuous judgment is considered to have high accuracy and real-time performance. When there is a risk of collision between multiple types of machines, the collision type and threshold established by the anti-collision model are used to determine the collision type and issue different types of alarms.

[0054] S7: Response strategy. According to the requirements of the multi-machine operation specifications, when different types of alarms occur, the project site needs to collect timely and effective response strategies to eliminate the alarm information and ensure safe operation on site.

[0055] This application uses the construction site as a real-world application scenario, selects tower cranes, truck cranes, and crawler cranes as the main machine types, and considers the height difference of the operation to divide the operation types into same-level operation and height difference operation. Same-level operation means that each machine type operates at the same height, and there is a risk of collision between the booms. Height difference operation means that each machine type operates at different heights, and there is a risk of collision between the boom and the wire rope. The collision calculation of multiple machine types in cross-operation is essentially a spatial geometry problem. This application simplifies the above two collision situations into two-dimensional and three-dimensional spatial movements, respectively, to simplify the anti-collision calculation process of the tower crane.

[0056] Furthermore, in S2, building an anti-collision model includes the following steps: S21, collecting data from multiple models; S22, establishing an anti-collision model at the same level; S23, establishing an anti-collision model at a height difference.

[0057] First, the tower crane structure is appropriately simplified. The simplified model of the flat-arm tower crane is as follows: Figure 2 shown.

[0058] S21. Data collection of multiple machine types, including: hoisting visualization through the hoisting area camera on the hoisting equipment, hoisting equipment arm length data collection through the boom length sensor on the hoisting equipment, hoisting equipment vertical rotation angle data collection through the boom angle sensor on the hoisting equipment, positioning the on-site position of each machine type and measuring the boom rotation angle through the Beidou positioning module on the hoisting equipment, hook rope quality monitoring through the winch mechanism camera on the hoisting equipment, hook length detection through the hook height sensor on the hoisting equipment, and data processing and calculation through the host on the hoisting equipment; wherein, hoisting equipment includes tower cranes and mobile lifting machinery.

[0059] Data collection of multiple types of machines, specifically, data collection of mobile lifting machinery selects the equipment shown in Table 1 for key data collection to further meet the needs of subsequent modeling, and uses the Beidou positioning module to realize real-time positioning of mobile lifting machinery and measurement of horizontal rotation angles.

[0060] Table 1 Hardware equipment list

[0061]

[0062]

[0063] S22. Establish a collision avoidance model at the same level, using the following method:

[0064] Select a point O on the ground of the construction site as the origin to establish the global reference coordinate system {O-xyz}, with the east as the x-axis, the north as the y-axis, and the vertical direction as the z-axis. The position matrix of the crane can be derived. Assume that two adjacent tower cranes TJ1 and TJ2 are arranged on the construction site, and the booms are operating at the same horizontal height. The two adjacent tower cranes are distributed as follows: Figure 3 As shown, TJ1 and TJ2 represent two adjacent tower cranes, and R1 and R2 are the boom lengths of tower cranes TJ1 and TJ2;

[0065] In this case, the boom has the longest radius, and its outer edge is the front end where the collision occurs. Assuming that the boom lengths of TJ1 and TJ2 are R1 and R2 respectively, when O1O2>R1+R2, there is no risk of cross-operation between the booms. When O1O2≤R1+R2, there is a cross-operation area T. Then, when the two tower crane booms move to the cross-operation area, there is a risk of collision. According to the set anti-collision threshold, an anti-collision warning is issued. This anti-collision model is also applicable when the tower crane and the mobile lifting machinery are hoisting at the same level.

[0066] S23. Establish a height difference anti-collision model. For two tower cranes, the method used is:

[0067] Assume that two adjacent tower cranes TJ1 and TJ2 are arranged at the construction site, operating at different levels. Assume that TJ1 is a low-level tower crane and TJ2 is a high-level tower crane. The two adjacent tower cranes are distributed as follows: Figure 4 As shown, H1 and H2 are the heights of tower cranes TJ1 and TJ2, R1 and R2 are the boom lengths of tower cranes TJ1 and TJ2, L1 and L2 are the distances between the hooks of tower cranes TJ1 and TJ2 and their respective trolleys, and r1 and r2 are the luffing lengths of the trolleys of tower cranes TJ1 and TJ2;

[0068] At this time, there is no collision risk between the crane arms. Assume that the operating heights of TJ1 and TJ2 are H1 and H2 respectively, the hook lengths are L1 and L2 respectively, and the trolley amplitudes are r1 and r2 respectively. The collision risk between the wire rope and the low tower boom is mainly considered. The side views of the two tower cranes are as follows: Figure 5 shown.

[0069] When O1O2>R1+r2, there is no risk of collision between tower cranes; when O1O2≤R1+r2, there is an intersection area between the two tower cranes in the two-dimensional space when viewed from above, but when H2-L2>H1, there is a height difference in the cross operation, and there is also no risk of collision. When H2-L2≤H1, there is a cross operation area T, and there is a risk of collision when the hooks of the two tower cranes move to the intersection area. This anti-collision model is also applicable when tower cranes and mobile lifting machinery are hoisting at different levels.

[0070] When considering multiple machine types, take tower crane and truck crane as an example. Figure 6As shown in the figure, assuming that the height of the mobile crane is H2, the arm length is R2, the horizontal rotation angle is θ, the tower crane is in a low position, and the mobile crane is in a high position, when O1O2>R1+R2cosθ, there is no collision risk. When O1O2≤R1+R2cosθ, the tower crane and the mobile crane have an intersection area in the two-dimensional space of the top view, but when H2+R2sinθ-L2>H1, there is a height difference in the cross operation, and there is no collision risk. When H2+R2sinθ-L2≤H1, there is a cross operation area T, and there is a collision risk when the tower crane and the mobile crane hook operate to the intersection area.

[0071] Similarly, the collision risk can be analyzed for cross operations in other situations. When the two tower cranes are operating at different levels, it can be regarded as a special case of θ = 0 degrees. When the two tower cranes are operating at the same level, it can be regarded as a special case of θ = 0 and H1 = H2.

[0072] In such Figure 6 Based on the operation model of various types of lifting equipment shown in the figure, the selection of warning and alarm thresholds is considered in combination with the braking process of the lifting equipment. Without considering external interference such as wind speed and load, it is assumed that the mechanical braking rotation angle of the tower crane or mobile crane is w and the braking distance is L. When braking, the variable frequency speed control method is adopted. It is assumed that there are n gears in total, and the braking rotation angles under each gear are w i , then w=∑w i , i=1...n, where the braking rotation angle in each gear is related to factors such as the state, model, and load of the crane type, and can be calculated based on actual equipment parameters, and the anti-collision warning and alarm thresholds can be set in the system.

[0073] by Figure 7 Taking the model as an example, the anti-collision threshold of the tower crane and the truck crane is calculated. The tower crane is on the left and the truck crane is on the right (at a high position). The truck crane runs counterclockwise. There is a collision risk between the lifting wire rope of the truck crane and the tower crane boom. Then L = 2R2sin(∑w i / 2), the alarm distance can be set in the system as, L 报 ≥L, the warning distance is magnified by a certain coefficient based on the actual situation on site, which can be set to For the amplification system, L 报 , There is no relevant provision in the current technical specifications. It is necessary to combine the actual conditions such as on-site wind speed and hanging load, and according to the angle of the anti-collision cross area of ​​the on-site lifting equipment, take appropriate values ​​and set them in the system. When considering that the tower cranes are also running relative to or in opposite directions, only the relative rotation angle between the two lifting equipments needs to be considered, that is, w i is the relative rotation angle.

[0074] Furthermore, the anti-collision system architecture of the present invention is designed as follows:

[0075] It consists of four parts: anti-collision server, positioning base station, gateway and data acquisition equipment. The data acquisition equipment varies slightly depending on the model. Figure 8 shown.

[0076] 1. Locate base station

[0077] The Beidou system + ground positioning auxiliary base station is used to provide the same coordinate system and horizontal reference plane for all multiple models entering the site for operation, providing basic conditions for multiple models to avoid collision. At the same time, the Beidou positioning system provides multiple models with high-precision positioning coordinates, altitude, and rotation angle in real time, avoiding errors caused by manual input of multiple models' heights when the heights of multiple models change. [8] , providing reliable data support for the anti-collision system.

[0078] 2. Anti-collision server

[0079] (1) Server large model calculation: Digital twin technology is used to realize trend analysis such as early warning and alarm for anti-collision. All multi-machine equipment uses three-dimensional parametric modeling to realize data visualization. The data collected by the equipment drives the three-dimensional model. The three-dimensional model runs synchronously with the on-site equipment, and the equipment status is analyzed in real time. The server large model calculates and analyzes the working conditions of all multi-machine types on site in real time, and early warnings are given in real time, and early warning data is transmitted to multiple machines on site.

[0080] (2) Data visualization based on web technology and digital twin technology: The data collected by the equipment drives the 1:1 three-dimensional model in the digital twin system in real time, and supervisors can view the operating conditions of multiple models consistent with the on-site situation in real time.

[0081] (3) The large and small digital twin models of the server and the terminal, synchronize redundant calculations, and mutually verify the alarm data of multiple models to ensure the accuracy of the anti-collision system data and the consistency of the data viewed by the driver and supervisor.

[0082] (4) The multiple aircraft types on site were restored using real 1:1 3D modeling technology. The booms, hook positions, and chassis information of the multiple aircraft types were all realistically displayed in the 3D model. The latitude and longitude data, amplitude, altitude, heading angle, and other attitude data of the multiple aircraft types collected on site were used to realistically restore the on-site positions and actual attitudes of the multiple aircraft types.

[0083] (5) The three-dimensional anti-collision algorithm based on digital twin technology actually takes into account the shapes of the boom, chassis, hook, etc. of various models. The algorithm is more accurate and the alarm is more precise.

[0084] (6) With the help of digital twin technology, the actual situation on site can be visualized; historical data can be visualized; even non-professionals can understand the on-site situation, effectively improving supervision efficiency.

[0085] 3. Gateway

[0086] The gateway is a data transmission medium that provides effective and reliable services for the collected data.

[0087] 4. Data acquisition equipment

[0088] There are slight differences depending on the machine model, and it mainly completes four tasks: device positioning, device posture collection, small model anti-collision calculation, and human-computer interaction.

[0089] (1) Equipment positioning: Accurately locate the longitude and latitude coordinates, altitude, and heading angle to provide a reference point for the entire system.

[0090] (2) Data collection: The height, amplitude, rotation, working condition and other posture data of multiple models are collected during operation and transmitted to the anti-collision server after processing.

[0091] (3) Small model anti-collision calculation: The digital twin small model technology realizes the anti-collision calculation between the local device and the surrounding equipment. The terminal device small model calculates and analyzes the working conditions of multiple types of machines associated with the local device, promptly issues sound and light alarms to the driver, and pushes data to the server for interaction.

[0092] (4) Large and small digital twin models, synchronized redundant calculations, to ensure the accuracy of the anti-collision system data.

[0093] (5) Human-machine interaction: Anti-collision warning alarm interaction. Multiple types of lifting equipment are operating simultaneously. When there is a risk of collision, an alarm message can be sent to the cab autonomously.

[0094] This system uses digital twin technology and Beidou RTK high-precision positioning technology to achieve everything from equipment data collection and processing, to three-dimensional modeling of multiple models, data-driven models, to real-time calculation and analysis, prediction of collision warning and alarm conditions between multiple models, and then to interface visualization. The posture data of multiple models such as hook height, amplitude, rotation, etc. collected on-site drive the equipment model in the twin system in real time. The equipment warning and alarm in the digital twin system are linked to the on-site equipment in real time, and the two sides are redundantly checked to ensure that the digital twin system displays and warns correctly. The collision warning, alarm, and no alarm status of the equipment are presented in different colors on the visual interface, and are accompanied by voice alarms.

[0095] In combination with on-site dispatch requirements and relying on system data, a large-screen display interface is established. The large-screen display interface is displayed in 2D and 3D interfaces respectively. When there is a collision warning or alarm, the collision warning and collision alarm voice broadcast are carried out.

[0096] In order to further explore the advancement of the anti-collision model proposed in this application, the mainstream anti-collision products in the market were investigated for effect comparison and analysis, as shown in Table 2. The collision dimension, height, position calibration method, diversity of research models, accident analysis method and SMS push were selected for comprehensive comparative analysis.

[0097] Table 2 Comparative analysis of the effects of each system

[0098]

[0099] First, in terms of collision dimension, external products mostly use two-dimensional anti-collision technology, while this application innovatively proposes a three-dimensional obstacle avoidance algorithm based on digital twins. Secondly, in terms of height and position calibration methods, compared with previous manual calibration, this application uses Beidou technology to automatically collect height and position data, which effectively avoids manual processing errors and is more efficient and real-time. Thirdly, this application integrates multiple models for consideration. Compared with previous studies that only considered the anti-collision between tower cranes, the model method of this application is more in line with the actual situation on site. Finally, in terms of accident restoration analysis, the model method of this application can be based on data twins for 3D playback, which is more intuitive. At the same time, it supports real-time information push of SMS APP when anti-collision occurs, and more efficient information flow helps to handle accidents in a timely manner.

[0100] In view of the shortcomings of traditional multi-machine anti-collision technology in terms of intelligence, real-time and rapidity, this application combines digital twin technology and Beidou satellite RTK technology fusion algorithm to build a multi-machine mobile anti-collision system that can realize multi-party real-time supervision, anti-collision, real-time alarm, and data storage and recording functions when multi-machine lifting equipment is operating, achieving real-time consistency between information flow and business flow. The following main conclusions were obtained:

[0101] (1) Realize on-site 3D visualization, greatly improving supervision efficiency. The equipment in the digital twin system corresponds to the location and action of on-site equipment one by one, meeting the needs of real-time positioning of mobile targets and further improving on-site supervision efficiency.

[0102] (2) Redundant verification and synchronous alarm effectively avoid blind spots in supervision. The anti-collision server implements redundant verification of pre-alarm information, and the server and the driver can issue alarms synchronously to avoid blind spots in supervision.

[0103] (3) The digital twin solution is used to ensure the accuracy of the results. The crane corresponds 1:1 to the actual model in the digital twin system, and there is no need to enlarge the crane boom range, making data calculation more realistic. The large model of the anti-collision server and the small model of the data acquisition device perform synchronous calculations and mutual verification, further ensuring the accuracy of the results.

[0104] In summary, this application combines digital twin technology to build a multi-machine lifting equipment anti-collision system, which can not only be used on construction sites, but also be extended to other engineering machinery fields. In addition, digital twin technology and Beidou satellite RTK technology can also be used for real-time progress monitoring on construction sites, foundation pit slope monitoring, labor distribution monitoring, etc.

[0105] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A construction site hoisting anti-collision method based on digital twin, characterized in that: The steps include: S1. Analysis of collision problems of multiple types of machines: During the construction process, the main factors and data parameters affecting the occurrence of collisions are determined by consulting literature and analyzing specific cases; S2. Establishing an anti-collision model and threshold: Building an anti-collision model covering all scenarios and setting a threshold based on the main collision scenarios and data parameters affecting the collision obtained in S1; S3. Digital twin model establishment: Collect key data based on the current technical background and the problem to be studied to map it into a virtual model; S4, Model and Site Interaction: After the digital twin data model of multiple models is established, the data of physical entities are continuously collected to achieve model calibration and timely status update; S5. Continuity judgment: judge the error between the physical entity and the digital twin model. When the error is greater than the threshold, the state of the model is considered discontinuous, and further data needs to be extracted to train the model to integrate key influencing features and associations; When the error is less than the threshold, the anti-collision model established based on the real-time data collected by the physical device and the current state of the virtual model predicts the possibility of a collision; S6. Determine the alarm type: The model judged by continuity is considered to have high accuracy and real-time performance. When there is a risk of collision between multiple types of machines, the collision type and threshold established by the anti-collision model are used to determine the collision type and issue different types of alarms. S7. Response strategy: According to the requirements of the multi-machine operation specifications, when different types of alarms occur, the project site adopts timely and effective response strategies to eliminate the alarm information; In S2, building the anti-collision model includes the following steps: S21. Data collection of multiple types of equipment: including hoisting visualization through the hoisting area camera on the hoisting equipment, data collection of the arm length of the hoisting equipment through the arm length sensor on the hoisting equipment, data collection of the vertical rotation angle of the hoisting equipment through the arm angle sensor on the hoisting equipment, positioning the on-site position of each type of equipment and measuring the rotation angle of the boom through the Beidou positioning module on the hoisting equipment, quality monitoring of the hook rope through the hoisting mechanism camera on the hoisting equipment, hook length detection through the hook height sensor on the hoisting equipment, and data processing and calculation through the host on the hoisting equipment; among which, the hoisting equipment includes tower cranes and mobile lifting machinery; S22, establishing a collision avoidance model at the same level; S23. Establish a height difference anti-collision model. For tower cranes and truck cranes, the method used is: Assume that the height of the mobile crane is H2, the arm length is R2, the horizontal rotation angle is θ, the tower crane is in a low position, and the mobile crane is in a high position. When O1O2>R1+R2cosθ, there is no collision risk. When O1O2≤R1+R2cosθ, there is an intersection area between the tower crane and the truck crane in the two-dimensional space viewed from above, but when H2+R2sinθ-L2>H1, there is a height difference in the cross-operation, and there is also no collision risk. When H2+R2sinθ-L2≤H1, there is a cross-operation area T, and there is a collision risk when the tower crane and the truck crane hook operate to the intersection area.

2. According to a construction site hoisting and lifting anti-collision method based on digital twinning according to claim 1, it is characterized in that: S22, establishing a same-level height anti-collision model, the method used is: Select a point O on the ground of the construction site as the origin to establish the global reference coordinate system {O-xyz}, with the east as the x-axis, the north as the y-axis, and the vertical direction as the z-axis, and derive the position matrix of the lifting machinery; Assume that two adjacent tower cranes TJ1 and TJ2 are arranged at the construction site, and the booms are operating at the same level. In this case, the boom has the longest radius, and its outer edge is the front end where the collision occurs; Assuming that the boom lengths of TJ1 and TJ2 are R1 and R2 respectively, when O1O2>R1+R2, there is no risk of cross operation between the booms; When O1O2≤R1+R2, there is an intersecting operation area T. When the two tower crane arms move to the intersecting area, there is a risk of collision. According to the set anti-collision threshold, an anti-collision warning is issued. This anti-collision model is also applicable when the tower crane and the mobile lifting machinery are performing hoisting operations at the same horizontal height.

3. According to the digital twin-based construction site hoisting anti-collision method of claim 1, it is characterized in that: The method used in S23, establishing a height difference anti-collision model for two tower cranes, is as follows: Assume that two adjacent tower cranes TJ1 and TJ2 are arranged at the construction site, operating at different levels. Assume that TJ1 is a low-level tower crane and TJ2 is a high-level tower crane; There is no collision risk for each boom. Assume that the operating heights of TJ1 and TJ2 are H1 and H2 respectively, the hook lengths are L1 and L2 respectively, and the trolley amplitudes are r1 and r2 respectively; When O1O2>R1+r2, there is no risk of collision between tower cranes; when O1O2≤R1+r2, there is an intersection area between the two tower cranes in the two-dimensional space when viewed from above, but when H2-L2>H1, there is a height difference in the cross operation, and there is also no risk of collision. When H2-L2≤H1, there is a cross operation area T, and there is a risk of collision when the hooks of the two tower cranes move to the intersection area. This anti-collision model is also applicable when tower cranes and mobile lifting machinery are performing hoisting operations at different levels.

4. The construction site hoisting and lifting anti-collision method based on digital twin according to claim 1 is characterized in that: The thresholds include anti-collision warning and alarm thresholds. Based on the operation model of various types of lifting equipment, combined with the braking process of the lifting equipment, the selection of warning and alarm thresholds is considered, without considering the wind speed, load and external interference. Assuming that the tower crane or mobile lifting mechanical braking rotation angle is w, the braking distance is L, and when braking, the variable frequency speed control method is adopted. Assuming that there are n gears in total, the braking rotation angles under each gear are w i , then w=∑w i , i=1…n, where the braking rotation angle in each gear is related to the state, model and load factors of the crane type. It is calculated based on the actual equipment parameters, and the anti-collision warning and alarm thresholds are set in the system.

5. A construction site hoisting and lifting anti-collision method based on digital twin according to claim 4, characterized in that: When calculating the anti-collision threshold between the tower crane and the truck crane, the truck crane runs counterclockwise, and there is a risk of collision between the truck crane's lifting wire rope and the tower crane's lifting arm. Then L = 2R2sin(∑w i / 2), the alarm distance is set in the system as, L 报 ≥L, the warning distance is enlarged by a certain coefficient based on the actual situation on site, and is set to is the magnification factor.

Images