Search

CN-122023472-A - Arch rib rotation construction monitoring device and method based on 4D point cloud and tension information

CN122023472ACN 122023472 ACN122023472 ACN 122023472ACN-122023472-A

Abstract

The invention relates to an arch rib rotation construction monitoring device and method based on 4D point cloud and tension information. The arch rib rotation is guided by a traction rope, the monitoring method comprises the steps of arranging a plurality of 4D laser radars in a visual field coverage field, installing a tension sensor on the traction rope, registering 4D point clouds obtained by the 4D laser radars to a unified coordinate system to obtain initial 4D point clouds, enabling the initial 4D point clouds to be in initial semantic registration with a BIM model before rotating the arch rib, enabling the 4D laser radars to obtain real-time 4D point clouds when rotating the arch rib, registering real-time 4D point clouds with the BIM model in real-time based on initial semantic registration, obtaining real-time data of the rotating arch rib based on the real-time 4D point clouds and the real-time semantic registration, executing at least one of early warning and guiding traction mechanisms if the real-time data exceeds a threshold value, acquiring the tension applied to the traction rope in real-time when rotating the arch rib, comparing the real-time acquired tension with a simulated tension value, and executing at least one of early warning and stopping the traction mechanism if the difference value exceeds the tension threshold value.

Inventors

  • XUE FAN
  • LIANG DONG
  • YOU KE

Assignees

  • 香港大学

Dates

Publication Date
20260512
Application Date
20241108

Claims (10)

  1. 1. The arch rib rotation construction monitoring method based on 4D point cloud and tension information, wherein the rotation of the arch rib is guided by a traction rope of a traction mechanism, and the monitoring method is characterized by comprising the following steps of: A plurality of 4D laser radars with visual fields covering a construction site are arranged; a tension sensor is arranged on the traction rope; registering the 4D point clouds obtained by the plurality of 4D laser radars in the construction site under a unified coordinate system to obtain an initial 4D point cloud of the construction site; before rotating the arch rib, carrying out initial semantic registration on the initial 4D point cloud and a pre-designed BIM model; During rotation of the arch rib, the plurality of 4D laser radars scan a construction site in real time and acquire a real-time 4D point cloud based on the unified coordinate system; based on the initial semantic registration, performing real-time semantic registration on the real-time 4D point cloud and the pre-designed BIM model; Based on the real-time 4D point cloud and real-time semantic registration, acquiring real-time state data of the arch rib during rotation, and if the real-time state number exceeds a corresponding threshold value, executing at least one action of early warning and guiding a traction mechanism; Acquiring a pulling force applied to the traction rope in real time through the pulling force sensor during rotation of the arch rib; and comparing the pulling force obtained in real time with the simulated pulling force value, and executing at least one action of early warning and stopping the traction mechanism if the difference exceeds the pulling force threshold value.
  2. 2. The method for monitoring the rotation construction of the arch rib based on the 4D point cloud and the tension information according to claim 1, wherein the step of registering the 4D point clouds obtained from the plurality of 4D lidar scanning construction sites to a unified coordinate system comprises: acquiring common features in the coincident fields of view of the plurality of 4D lidars; Constructing a reference coordinate system by using a 4D point cloud of one of the 4D laser radars; calculating registration parameters of the 4D point clouds of the rest 4D laser radars with respect to the reference coordinate system based on the common features; And uniformly registering the 4D point clouds of the rest 4D laser radars under the reference coordinate system based on the registration parameters.
  3. 3. The method for monitoring rib rotation construction based on 4D point cloud and tension information according to claim 2, wherein the step of obtaining common features in the overlapping fields of view of the plurality of 4D lidars comprises: when the common characteristics cannot be obtained, arranging the reflecting plate in the overlapped view as the common characteristics; Obtaining relevant point clouds of the reflector in the 4D point clouds of each 4D laser radar based on intensity filtering; Constructing a reference coordinate system by using a 4D point cloud of one of the 4D laser radars; Calculating registration parameters of the 4D point clouds of the rest 4D laser radars with respect to the reference coordinate system based on the reflector-related point clouds in the 4D point clouds of the rest 4D laser radars; And uniformly registering the 4D point clouds of the rest 4D laser radars under the reference coordinate system based on the registration parameters.
  4. 4. The method for monitoring rib rotation construction based on 4D point cloud and tension information according to claim 1, wherein the step of performing initial semantic registration of the initial 4D point cloud with a pre-designed BIM model comprises: Downsampling the BIM model into a virtual point cloud model; and carrying out semantic registration on the virtual point cloud model and the initial 4D point cloud.
  5. 5. The method for monitoring the rotation construction of the arch rib based on the 4D point cloud and the tension information according to claim 1, wherein after the plurality of 4D lidars scan the construction site in real time and acquire the real-time 4D point cloud based on the unified coordinate system, the method further comprises the following steps: Transmitting the real-time 4D point cloud to a remote processor through a wireless network according to a fixed sampling frequency, wherein the real-time 4D point cloud is compressed at a transmitting end, and the real-time 4D point cloud is decompressed at a receiving end.
  6. 6. The method of claim 1, wherein the pre-designed BIM model includes a plurality of BIM sub-stage models for a plurality of sub-stages of the pre-design of the rotational construction, and the step of real-time semantic registering the real-time 4D point cloud with the pre-designed BIM model includes: registering the real-time 4D point cloud with a related BIM sub-phase model.
  7. 7. The method for monitoring arch rib rotation construction based on 4D point cloud and tension information according to claim 1, wherein the real-time semantic registration result of the next frame based on the time sequence is a local adjustment of the real-time semantic registration result of the last frame.
  8. 8. The method for monitoring the rotation construction of the arch rib based on the 4D point cloud and the tensile force information according to claim 1, wherein the step of acquiring real-time state data of the arch rib during rotation based on the real-time 4D point cloud and the real-time semantic registration result comprises at least one of the following: Comparing the real-time 4D point cloud with the initial 4D point cloud to obtain real-time deformation data of the arch rib about different deformation types; comparing the related pre-designed BIM sub-stage models in the BIM model based on the initial semantic registration result and the real-time semantic registration result to obtain difference data of the real-time three-dimensional position of the arch rib and the target three-dimensional position; and analyzing the real-time 4D point cloud to obtain the real-time rotation speed of the arch rib.
  9. 9. The method for monitoring rib rotation construction based on 4D point cloud and tension information according to claim 1, wherein the step of comparing the tension acquired in real time with the simulated tension value comprises: extracting a related simulated tension value from a pre-designed BIM model based on real-time state data of the arch rib during rotation; and comparing the pulling force obtained in real time with the simulated pulling force value.
  10. 10. A monitoring device for rib rotation construction based on 4D point cloud and tension information, for performing the monitoring method of any one of claims 1 to 9, the monitoring device comprising: The field acquisition module comprises a plurality of 4D laser radars and a tension sensor; the communication module is connected to the field acquisition module and is configured to transmit data obtained by the field acquisition module to the processor; The mapping module is configured to register the 4D point clouds obtained by the plurality of 4D laser radars in the construction site under a unified coordinate system, and perform semantic registration on the 4D point clouds scanned in the construction site and a pre-designed BIM model; the calculation analysis module is configured to calculate whether real-time state data of the arch rib exceeds a corresponding threshold value during rotation and whether real-time tension of the arch rib exceeds a preset tension threshold value based on the data acquired by the field acquisition module and registration of the mapping module; And the control module is configured to send out at least one instruction of the two instructions of early warning and guiding the traction mechanism when the real-time state data exceeds the corresponding threshold value, and send out at least one instruction of the two instructions of early warning and stopping the traction mechanism when the real-time tension exceeds the tension threshold value.

Description

Arch rib rotation construction monitoring device and method based on 4D point cloud and tension information Technical Field The invention relates to the field of early warning in the bridge rotation construction process, in particular to a device and a method for monitoring arch rib rotation construction based on 4D point cloud and tension information. Background In the construction process of the steel arch rib of the existing large-span steel arch bridge exceeding one hundred meters, one widely adopted construction method is to assemble the steel arch rib on the shore and then move the assembled structure in place through horizontal rotation or vertical rotation. Advantages of this construction method include no need for large land and water lifting machinery, the ability to accommodate different river bank and river depth conditions, and reduced safety risks due to lifting instability. Proved by a large number of practices, the preparation method has the advantages that, the construction method for rotating the assembled shore can reduce the construction cost and accelerate the construction speed. The steel ribs are inevitably displaced from the intended position during rotation. The traditional steel arch bridge construction early warning and deviation correcting are mainly based on analysis of manual measurement results in the rotation process. In general, some marking points, such as prisms or reference targets, need to be installed at key points of the steel arch rib to be rotated in advance. In the rotation process, constructors measure the pre-installed identification points through measuring equipment such as a total station and the like, and then analyze the measured data, so that whether the steel arch rib rotation process is carried out according to an expected plan or not and whether abnormal deformation occurs or not are determined. The method can only observe the displacement of the fixed point where the identification point is located, and meanwhile, the problems of delay, inaccuracy and the like exist in information transfer between the field manual observation data and the command center, so that the construction process is difficult to monitor in real time, and the potential safety hazard is difficult to discover and process in time. In recent years, a large number of monitoring systems based on machine vision are presented, and the monitoring systems have the advantages of non-contact, long distance, automation and the like. However, these monitoring systems can only measure relative displacement in two-dimensional planes at limited marking points in the structure, and have difficulty in identifying the three-dimensional spatial position of the steel structural arch rib. In addition, the accuracy of these monitoring systems is also difficult to meet the requirements of accurate construction. For example, published Chinese patent application CN116437043A of 2023, 7 and 14 discloses a system and a method for monitoring a lifting video of a bridge girder. The invention acquires the position information of the monitored object by utilizing a high-precision Beidou positioning technology, and calculates the direction and the angle of the camera to be rotated according to the position information before and after the displacement of the monitored object by the monitoring center, thereby controlling the camera to track the monitored object in real time for monitoring. The monitoring system has the defect that the plane position movement sensing in the visual field range can be realized, and the three-dimensional process of the construction movement cannot be monitored well. Some known engineering cases adopt a three-dimensional scanning technology to position components for site construction, but the traditional three-dimensional scanning technology has the defect of long time consumption of single scanning, so that the construction process is required to cooperate with monitoring operation, which reduces the construction efficiency and increases the complexity of construction organization. For example, published Chinese patent application CN115491984A of 2022, 12 and 20 discloses a device and a method for hoisting a large-span steel box girder based on three-dimensional laser scanning. The invention realizes the accurate installation of the large-span steel box girder by means of the digital scanning and modeling of the three-dimensional laser scanning system and by means of the positioning mechanism, can greatly reduce the installation time of the steel box girder and remarkably improve the installation precision. The three-dimensional point cloud scanning method has the defects that one-time scanning can be completed only by performing long-time startup scanning on a static scene, and the scanning mode needs to suspend the construction process to cooperate, so that the construction progress is greatly reduced, and meanwhile, the complexity of construction organization is increased. Disclosur