A multi-source perception fusion iron bridge swivel real-time monitoring and early warning method

Abstract

The application discloses a kind of multi-source perception fusion's iron bridge involved real-time monitoring and early warning method of swivel, comprising: S1, generating the real scene three-dimensional model of construction site;S2, the real scene three-dimensional model is loaded in three-dimensional scene, and the basic parameters required for monitoring and early warning are extracted;S3, the real scene three-dimensional model is analyzed to the bridge deck area involved in iron bridge, and the installation position for deploying beidou positioning equipment and attitude sensor is selected;S4, the monitoring data collected is parsed, and the key index value of swivel monitoring of iron bridge involved is obtained;Data fusion is carried out to monitoring result, and the real-time attitude monitoring result of bridge swivel is obtained;S5, the real-time coordinates of four corner points of bridge girder are calculated, whether the intersection exists between the profile of iron bridge involved and railway clearance is monitored, when intersection, the clearance of swivel girder to catenary pole top is calculated;Swivel angle-time simulation curve is calculated;S6, the structural state change in swivel process is restored, and abnormal working condition is identified and early warned in real time.

Description

Technical Field

[0001] This invention relates to the field of bridge rotation construction technology, specifically to a real-time monitoring and early warning method for railway bridge rotation based on multi-source sensing fusion. Background Technology

[0002] Bridge rotation technology, as a key construction method, is widely used in various highway and railway projects, especially in high-density operating environments such as crossing existing railway lines, where it possesses irreplaceable advantages. This technology involves prefabricating the bridge structure at a predetermined location and then using a rotation device to rotate the entire bridge to the designed orientation. This effectively avoids the problem of prolonged closure of existing railway lines, improves construction efficiency, reduces interference with railway operations, and yields significant economic and social benefits.

[0003] However, bridge rotation construction is inherently characterized by "high precision and high risk." This risk is particularly prominent in railway-related projects: due to the dense operation and frequent traffic on existing railway lines, any deviation in posture or positioning during the rotation process can directly threaten train safety, with extremely serious consequences. Therefore, the construction of railway-related bridge rotation places more stringent requirements on the monitoring and control of the entire process. It not only needs to achieve high precision and high efficiency, but also must have strong real-time response capabilities and high early warning reliability to ensure safe and controllable construction.

[0004] Current bridge rotation construction monitoring methods have the following drawbacks:

[0005] 1. Traditional measurement methods rely on equipment such as total stations to collect discrete points manually, which has two major technical drawbacks: First, the location of monitoring points changes constantly during the bridge rotation process, making it difficult to capture accurately in real time using a total station; second, the labor cost is high, as personnel are required to be on duty and continuously measure on both sides of the rotating beam throughout the process.

[0006] 2. The GNSS-based monitoring method obtains the coordinates of the beam and calculates its attitude information through GNSS equipment. Although it has the advantages of strong real-time performance and high degree of automation, this method still has two problems: First, the installation points of GNSS equipment are mostly determined by manual estimation and steel tape measurement, which is easily affected by construction errors, leading to point selection deviations and thus affecting monitoring accuracy; Second, GNSS positioning relies on radio carrier signals, which are easily affected by external environmental interference such as the ionosphere, and its stability needs to be improved.

[0007] In summary, the construction of railway bridges involving rotation places more stringent demands on the accuracy, real-time performance, and early warning capabilities of monitoring systems. Existing monitoring technologies each have their shortcomings and are insufficient to fully meet the "high reliability and zero fault tolerance" safety control requirements of railway bridge rotation construction. Summary of the Invention

[0008] To address the problems existing in the prior art, this invention provides a real-time monitoring and early warning method for the rotation of railway bridges using multi-source sensing fusion, which is highly efficient, accurate, stable, and low-cost. This method enables real-time automated monitoring and early warning of the bridge rotation construction process.

[0009] Therefore, the present invention adopts the following technical solution:

[0010] A method for real-time monitoring and early warning of railway bridge rotation involving multi-source sensing fusion, comprising the following steps:

[0011] This invention constructs an improved monitoring method that integrates precise deployment, stable positioning, and intelligent early warning. By optimizing equipment selection methods, improving deployment accuracy, and strengthening signal anti-interference capabilities, while introducing an intelligent early warning mechanism, it effectively enhances the ability to identify and respond to anomalies during bridge rotation.

[0012] Compared with the prior art, the present invention has the following beneficial effects:

[0013] 1. Compared with the original manual measurement method, the method of the present invention can automatically extract the basic parameters of bridge rotation and collect data during the bridge rotation construction process, which greatly saves manpower and effectively improves the real-time monitoring of rotation.

[0014] 2. The method of the present invention combines BeiDou positioning and attitude sensor data for monitoring, which effectively improves the monitoring accuracy and stability under different environmental conditions compared to GNSS, a single monitoring method.

[0015] 3. This invention proposes a method for selecting and installing monitoring equipment based on a real-world 3D model, which effectively improves the efficiency of site selection and installation compared to on-site surveying.

[0016] 4. The method of the present invention monitors in real time the rotation angular velocity and time, beam tilt and over-rotation, the entry of the rotating beam into the railway clearance, and the clearance between the rotating beam and the contact wire pole. This enables timely early warning of problems during the rotation process, improves construction safety, and reduces potential economic losses. Attached Figure Description

[0017] Figure 1 This is a flowchart of the real-time monitoring and early warning method for the rotation of railway bridges according to the present invention;

[0018] Figure 2 This is a schematic diagram of the rotating beam entering the railway clearance in this invention;

[0019] Figure 3 This is a schematic diagram of the installation of the Beidou positioning device and attitude sensor for a straight beam in an embodiment of the present invention;

[0020] Figure 4This is a schematic diagram of the rotation interface of the real-time monitoring system for the rotation of railway bridges according to the present invention;

[0021] Figure 5 This is an example of the rotation angle calculated by BeiDou and attitude sensors during the rotation of a straight beam in an embodiment of the present invention, as well as the change in rotation angle after data fusion.

[0022] Figure 6 This is a comparison chart of the actual rotation angle-time curve and the simulated rotation angle-time curves corresponding to the highest and lowest angular velocities in an embodiment of the present invention;

[0023] Figure 7 This is a schematic diagram showing the installation positions of the curved beam Beidou positioning device and attitude sensor in an embodiment of the present invention;

[0024] Figure 8 The embodiments of the present invention show the uncorrected Beidou positioning device curved beam attitude monitoring results and the corrected and fused curved beam attitude monitoring results. Detailed Implementation

[0025] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings and embodiments. Obviously, the following embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] See Figure 1 The multi-source sensing fusion method for real-time monitoring and early warning of railway bridge rotation of the present invention includes the following steps:

[0027] S1. Acquisition of a real-world 3D model of the railway bridge rotation construction site: Oblique photography was conducted by drone aerial flight, and a real-world 3D model of the railway bridge rotation construction site was generated based on the oblique photography images using 3D reconstruction.

[0028] During drone flight, the flight path should be parallel to the railway line, and the horizontal distance between its projection and the rails should be maintained at 25 to 30 meters. This distance is greater than the limit for safety control around the railway and ensures the clarity of the aerial imagery. The flight path must not fly over the railway line, and its altitude should be 20 to 30 meters higher than the rotating beam. The planar error of the real-scene 3D model based on oblique photography should be less than 2 centimeters, and the elevation error should be less than 5 centimeters.

[0029] S2, Extraction of monitoring parameters for railway bridge rotation based on real-scene 3D model: The real-scene 3D model obtained in S1 is loaded into the 3D scene, and the basic parameters required for real-time monitoring and early warning are extracted through manual measurement.

[0030] The basic parameters include: the geographic coordinates of the beam surface rotation center O point (lon). o lat o h o The initial coordinates of the four corner points of the beam {C, D, E, F}, and the sequence of railway line clearance points {P1, P2, ..., P} n}, the three-dimensional coordinate sequence of the top of the contact wire pole {G1, G2, ..., G} m The sampling point sequence for mileage and beam surface thickness at 5-meter intervals is (mileage1, thickness1), (mileage2, thickness2), ..., (mileage...). k thickness k ), bridge rotation direction, rotation angle α of the rotating beam total Beam length L beam Beam width W beam Beam height H beam .

[0031] S3, Site selection and installation of monitoring equipment based on real-scene 3D model: The real-scene 3D model is used to analyze the beam surface area of ​​the railway bridge to understand the distribution of fences, counterweights, structural components and other debris. On the basis of avoiding interference, a suitable installation location is selected in combination with the structural characteristics of the beam surface for the deployment of Beidou positioning equipment and attitude sensors.

[0032] The Beidou positioning device should have centimeter-level positioning accuracy and real-time positioning capability at frequencies of 1Hz or higher under unobstructed signal conditions. It should be able to continuously transmit location information back to the designated server and have good waterproof and dustproof capabilities as well as wide temperature adaptability, and be able to adapt to complex construction environments.

[0033] The installation locations of the BeiDou positioning devices should meet the following requirements: the BeiDou positioning devices should be installed on the centerline of the bridge; one BeiDou positioning device should be installed at each of the large and small mileage ends of the beam, and the distances from the installation points of the two BeiDou positioning devices to the rotation center of the beam surface should be equal; the devices should be installed on fixed supports, and the surrounding environment should be open and unobstructed to ensure the reception of high-quality BeiDou signals.

[0034] The attitude sensor should have an attitude monitoring accuracy better than 0.1°; it should also have an attitude data acquisition frequency higher than 1Hz and be able to upload the data to a designated server in real time; and it should meet outdoor use requirements such as waterproof and dustproof, and strong temperature adaptability.

[0035] The attitude sensor should be installed at the center of rotation of the beam surface. During installation, it should be ensured that its three axes are strictly parallel to the coordinate axes of the beam to ensure the accuracy of the attitude data.

[0036] According to the installation requirements of the BeiDou positioning device and attitude sensor, suitable points are selected in the real-world 3D model, and their 3D coordinate information is extracted. During the actual installation process, the extracted 3D coordinate information is measured using the BeiDou positioning device to obtain the actual position of the installation point on the beam surface, ensuring that the device is installed accurately in the predetermined position.

[0037] S4, BeiDou positioning and attitude awareness monitoring data processing and fusion:

[0038] According to the data transmission protocol between the BeiDou positioning device and the attitude sensor, the monitoring data collected by the BeiDou positioning device and the attitude sensor is analyzed, and key indicator values ​​for the rotation monitoring of railway bridges, including the azimuth angle, tilt angle, instantaneous angular velocity, and average angular velocity of the rotating beam, are obtained through data processing. To improve the accuracy of rotation monitoring and its stability under different environments, the monitoring results of the BeiDou positioning device and the attitude sensor are fused based on the mean square error to obtain high-precision real-time attitude monitoring results for bridge rotation, thus improving the accuracy of rotation monitoring. The specific steps of the data processing and fusion are as follows:

[0039] S41, BeiDou Positioning Equipment Monitoring Data Processing: The data packets transmitted by the BeiDou positioning equipment adopt the standard GPGGA format; the packet content includes key information such as equipment number, data measurement time, longitude, latitude, elevation, positioning type, and standard error. After obtaining the above-mentioned formatted data, coordinate transformation and rotation attitude calculation are performed. The specific processing steps are as follows:

[0040] S411, Coordinate Transformation: The geographic coordinate system result of BeiDou positioning is transformed to the construction coordinate system. The coordinate transformation from the geographic coordinate system to the construction coordinate system is calculated using a seven-parameter coordinate transformation model, as shown below:

[0041]

[0042] In the formula, X, Y, Z and X′, Y′, Z′ are the coordinates in the geographic coordinate system and the construction coordinate system, respectively; ΔX, ΔY, ΔZ are the translation parameters; ω x ω y ω z is the rotation parameter; s is the scale factor.

[0043] The translation parameters, rotation parameters, and scale factors are calculated using the least squares method based on three or more known control points.

[0044] S412, Rotation Attitude Calculation: Based on the coordinates of the transformed construction coordinate system obtained in S411, the azimuth angle, tilt angle, instantaneous angular velocity, and average angular velocity of the rotating beam are calculated. Data fusion of the monitoring results from both large and small mileage terminals is then performed using the positioning type and error of the BeiDou positioning equipment. Details are as follows:

[0045] The geographic coordinates of the beam rotation center O extracted in S2 are processed using the coordinate transformation model of S411 to obtain the construction coordinate system coordinates O′(x0,y0,z0) of the beam rotation center. These coordinates are then combined with the initial coordinates A0(x0,y0,z0) of the Beidou positioning devices at both ends of the beam. A0 ,y A0 ,z A0 ) and B0(x B0 ,y B0 ,z B0 The initial azimuth angle α0 of the beam is calculated as follows:

[0046]

[0047] Taking BeiDou positioning device A as an example, assume that the construction coordinates of the location of the BeiDou positioning device at time i are A. i (x Ai ,y Ai ,z Ai At this moment, the azimuth angle α of the rotating beam is... i And the rotation angle α of the rotating beam rotated The following formulas are used respectively:

[0048] α i =arctan((y Ai -y0) / (x Ai -x0))+α constA

[0049] α rotated =α i -α0

[0050] In the formula, α constA To correct the azimuth angle corresponding to the shape of the rotating beam to a constant, use the following formula:

[0051]

[0052] The initial tilt angle θ0 of the bridge, and the tilt angle θ at time i. i The formula for calculating the change in tilt angle Δθ is as follows:

[0053]

[0054] Δθ=θ i -θ0

[0055] Instantaneous angular velocity ω and average angular velocity of rotation The calculation formula is as follows:

[0056]

[0057] Two sets of rotation attitude monitoring results were obtained from two Beidou positioning devices installed at the large and small mileage ends of the beam. These results were then fused using the mean square error, resulting in:

[0058]

[0059] In the formula, These represent the attitude monitoring results from the fusion of data from two BeiDou positioning devices, the attitude monitoring results from BeiDou positioning device A, and the attitude monitoring results from BeiDou positioning device B, respectively; RMS A RMS B These represent the horizontal mean square errors of BeiDou positioning devices A and B, respectively; fix A fix B These represent the positioning types of BeiDou positioning devices A and B, respectively. The positioning type `fix` is defined as shown in the formula:

[0060]

[0061] S42, Attitude Sensor Monitoring Data Processing: The azimuth angle, tilt angle, instantaneous angular velocity, and average angular velocity of the rotating beam monitored by the attitude sensor are obtained through parsing the data message format of the attitude sensor. Valid attitude information parameters are extracted and used as the monitoring results of the attitude sensor. Specifically:

[0062] The monitoring data collected by the attitude sensor includes: the rotation angles of the three coordinate axes relative to the initial attitude of the device, the instantaneous rotational angular velocity of each axis, and the corresponding data acquisition timestamps. This data is transmitted wirelessly to a designated server in the form of TCP packets. Based on the data packet format and decoding rules, the received monitoring data collected by the attitude sensor is parsed and processed frame by frame to extract valid attitude information parameters. As a monitoring result from the attitude sensor.

[0063] S43, Monitoring Result Fusion: Based on the mean square error of the BeiDou positioning equipment and attitude sensors, the monitoring results of the two are weighted and fused to obtain high-precision real-time attitude monitoring results of the bridge rotation. Details are as follows:

[0064] Based on the attitude monitoring results from the BeiDou positioning device obtained from S412 and the attitude sensor obtained from S42, in order to achieve a comprehensive and accurate description of the spatial attitude of the bridge beam during its rotation, the data from these two types of monitoring results need to be fused. The fusion process first pairs the monitoring results from the BeiDou positioning device and the attitude sensor at the same time point based on the monitoring time, and then performs weighted fusion, as shown in the following formula:

[0065]

[0066] In the formula, p bd p sensor These are the weights of the monitoring results from the BeiDou positioning equipment and the attitude sensor, respectively, and their values ​​can be calculated from the mean square error (RMS) of the two types of monitoring results. The final attitude monitoring result is obtained by fusing the monitoring results from the BeiDou positioning equipment and the attitude sensor, where α final Let Δθ be the azimuth angle of the rotating beam at the current moment. final ω represents the change in the beam's inclination angle. final It is the instantaneous angular velocity. The mean angular velocity.

[0067] The mean square error of the BeiDou positioning device is significantly affected by the installation environment and signal strength; therefore, the returned data packets typically include a mean square error value calculated in real time. The mean square error of the attitude sensor is less affected by the environment and can be calculated using observations over a period of time after installation. The weights of the two types of monitoring results, calculated based on the mean square errors of both, are as follows:

[0068]

[0069] In the formula, RMS bd RMS sensor These are the mean square error values ​​of the measurement data from the BeiDou positioning equipment and the attitude sensor, respectively.

[0070] S5, Real-time Calculation of Monitoring Indicators: Based on the high-precision real-time attitude monitoring results of the bridge rotation obtained in S4, the monitoring indicators for the rotating beam entering the railway clearance, the clearance between the rotating beam and the contact wire pole, and the rotation angle-time simulation curve are calculated. The specific steps are as follows:

[0071] S51, Monitoring of the Rotating Beam Entering the Railway Clearance: Based on the real-time attitude monitoring results obtained in S43, the real-time coordinates of each corner point of the beam are calculated. The vector product is used to monitor whether the outline of the rotating beam intersects with the railway clearance, thereby determining whether the rotating beam has entered the airspace above the railway clearance. For example... Figure 2 As shown, the specific steps are as follows:

[0072] S511, based on the initial coordinates {C, D, E, F} of the four corner points of the beam extracted in S2, calculate their real-time coordinates using the following formula:

[0073]

[0074] In the formula, (x′, y′) are the real-time coordinates of the corner points of the beam; (x, y) are the initial coordinates of the corner points of the beam; (x0, y0) are the coordinates of the rotation center of the rotating beam; and α is the difference between the real-time azimuth angle and the initial azimuth angle of the beam.

[0075] S512, combined with S2, the railway clearance point sequence {P1, P2, ..., P...} is obtained. n The cross product is used to detect whether each side of the beam intersects with each section of the railway clearance. The formula is as follows:

[0076]

[0077] Taking CD side as an example, when there is a railway clearance that satisfies the above inequalities, the beam intersects with the railway clearance, indicating that the beam has entered the airspace above the railway clearance.

[0078] S52, Calculate the clearance from the rotating beam to the top of the contact wire pole:

[0079] Based on the fact that the S51 rotating beam has entered the airspace above the railway clearance, the half-plane method is used to determine whether the rotating beam has passed above the top of the contact wire pole:

[0080] First, based on the real-time three-dimensional coordinate data of the beam corner points obtained in S51 during the rotation process (the z-coordinate remains unchanged during rotation), combined with the coordinates of the top of the contact wire pole extracted in S2, the half-plane method is used in a unified coordinate system to determine whether the beam passes over the top of the contact wire pole. The half-plane method involves calculating the vector product of the edge vector and the vector from the contact wire pole to the edge starting point for each edge of the beam. If all the obtained vector products are in the same direction, it indicates that the beam has passed over the top of the contact wire pole.

[0081] The calculation steps for the monitoring index of the clearance between the rotating beam and the top of the contact wire pole are as follows:

[0082] S521, When the beam passes over the contact wire pole, the planar coordinates of the top of the contact wire pole are projected onto the centerline of the bridge to obtain the projection point G′(x). G′ y G′ , z G′ ) and the corresponding mileage m G′ The elevation at the projection point G′ of the beam surface can be obtained by interpolating the beam surface elevations at points A and B using the Beidou positioning device.

[0083]

[0084] In the formula, m A ,m B These are the mileages at the installation locations of Beidou positioning devices A and B, respectively; z A ,z B The elevation of the beam surface at points A and B of the Beidou positioning device.

[0085] S522, Elevation z of beam surface projection point G′ obtained from S521 G′ Elevation z extracted by S2 G (Obtained from the three-dimensional coordinate sequence of the top of the contact wire pole), and combined with the formula f(m) for the change of beam thickness with mileage, the net clearance Cl from the rotating beam to the top of the contact wire pole is as follows:

[0086] Cl=z G′ -f(m G′ )-z G

[0087] The formula f(m) for the change in beam thickness with mileage is derived from the sequence of mileage sampling points at 5-meter intervals obtained in S2: {(mileage1,thickness1),(mileage2,thickness2),…,(mileage…)}. k thickness k Based on the actual shape of the beam, the curve is obtained by piecewise function multi-curve fitting.

[0088] S53, Calculate the rotation angle-time simulation curve: In order to realize real-time monitoring and early warning of overtime, overspeed and other issues during the bridge rotation process, it is necessary to calculate the simulation curve of the beam angle changing with time during the rotation process, compare it with the rotation curve in the actual rotation construction process, and provide data support for the rotation construction control.

[0089] The rotation process includes four stages: acceleration, constant speed, deceleration, and fine adjustment.

[0090] The acceleration phase begins when the rotation window is opened and ends when the rotation angle reaches 5 degrees. This phase usually requires the deployment of safety personnel, power outage of the overhead contact line, and adjustment of the rotation equipment.

[0091] The uniform speed phase begins when the angle of rotation is 5 degrees and ends when the remaining angle is 5 degrees, during which the rotating beam maintains a basically uniform speed of rotation.

[0092] The deceleration phase begins when the remaining angle is 5 degrees and ends when the remaining angle is 1 degree, during which the rotation speed gradually decreases until it stops.

[0093] The fine-tuning phase begins when the remaining angle is 1 degree and ends when the rotating beam is fully in place, and the rotating beam is positioned through multiple inching operations.

[0094] Since the time required for the acceleration phase during actual rotation is uncertain, the rotation angle-time simulation curve is calculated according to the following formula when entering the uniform velocity phase:

[0095]

[0096] In the formula, α simu α represents the simulated rotation angles during the uniform and deceleration phases; t represents the rotation time; t1, t2, and t3 represent the start, start, and end times of the uniform rotation phase, respectively; ω′ represents the angular velocity during uniform rotation; α total The angle at which the rotating beam is to be rotated; a desc This refers to the angular acceleration during the deceleration phase.

[0097] Based on the formula, the curves of the rotation angle changing with time under the conditions of rotation construction at the highest and lowest angular velocities can be calculated, and thus an early warning can be given for actual rotation curves that exceed the speed range.

[0098] The maximum rotational angular velocity is calculated based on the cantilever length of the rotating beam:

[0099]

[0100] In the formula, ω max The maximum rotational angular velocity, in rad / min; L a This is the cantilever length of the rotating beam, which is half the total length of the rotating beam in a symmetrical rotating beam, and is expressed in meters (m).

[0101] Angular acceleration a during deceleration phase desc According to national and local standards, the value should be controlled at 4.2*10. -5 rad / s 2 the following.

[0102] For the minimum rotational angular velocity ω min First, calculate the expected end time of the deceleration phase based on the end time of the rotation window and the reserved time of the fine-tuning phase, and then solve it using the rotation angle-time simulation curve formula.

[0103] S6, Rotation Display and Early Warning: By constructing a 3D real-world construction site scene integrating BIM and GIS, and combining real-time monitoring results to drive dynamic updates of the rotating beam's posture, the structural state changes during the rotation process are fully reproduced. Simultaneously, based on the real-time bridge rotation posture monitoring results and indicators from S4 and S5, abnormal working conditions are identified and warned in real time. The specific steps of the rotation display and early warning are as follows:

[0104] S61, Construction of a 3D Scene of the Rotation Site Combining BIM and GIS: A 3D geographic environment of the rotation construction site is constructed on the web page. The real-scene 3D model obtained in S1 is modified and deployed into the geographic environment, and the BIM model of the rotating beam is superimposed on it. The model modification refers to removing the rotating beam and its auxiliary structures, various construction equipment, etc. from the real-scene 3D model.

[0105] S62, Parameter-driven real-time attitude update of the rotating beam: The real-time attitude monitoring results of the bridge rotation obtained in S4 are sent to the web page via WebSocket. The web page updates the horizontal rotation and tilt state of the rotating beam BIM model in real time according to the attitude monitoring results, so as to realize the synchronous rotation of the model and the beam.

[0106] S63, Warning for body rotation monitoring item:

[0107] Based on the real-time attitude monitoring results and indicators of bridge rotation obtained in S4 and S5, the rotation monitoring items are displayed and warnings are issued. These rotation monitoring items include: rotation angular velocity and time monitoring, beam tilt and over-rotation monitoring, monitoring of the rotating beam entering the railway clearance, and monitoring of the clearance between the rotating beam and the contact wire pole. Details are as follows:

[0108] The rotational angular velocity and time monitoring are obtained through S4, which shows the rotation angle α of the rotating beam. rotated Instantaneous angular velocity ω final Based on the rotation angle-time curve information obtained from S5, the following warnings are issued:

[0109]

[0110] In the formula, type alert Warning type; ω final ω is the instantaneous angular velocity. max ω min These are the highest and lowest rotational angular velocities in S53, respectively; α rotated The rotating beam has rotated by an angle; α simuMax α simuMin The simulated rotation angles are for the highest and lowest rotation angular velocities, respectively.

[0111] The beam tilt and over-rotation monitoring are shown below:

[0112]

[0113] In the formula, Δθ represents the change in the beam's inclination angle; Δθ threshold This is the threshold value for the change in the beam's tilt angle.

[0114] The monitoring of the rotating beam entering the railway clearance is based on the S51 rotating beam entering the railway clearance monitoring index. When the beam intersects with the railway clearance, a prompt is issued indicating that the rotating beam has entered the railway clearance.

[0115] The clearance monitoring from the rotating beam to the top of the contact wire pole is shown below:

[0116]

[0117] In the formula, Cl represents the clearance from the rotating beam to the top of the contact wire pole at the current moment; Cl min This is the net clearance threshold.

[0118] Example 1

[0119] This embodiment focuses on a rotating bridge in a railway engineering project. The bridge is a straight beam with a total length of 100 meters and a rotation angle of 71.6 degrees, including a trial rotation of 5 degrees and a final rotation of 66.6 degrees. The rotating beam is arranged symmetrically along the centerline, with the rotation center located at the geometric center of the beam.

[0120] During implementation, firstly, oblique images of the construction site are acquired via drone flight, and a realistic 3D model is constructed. Then, the realistic 3D model is loaded into the 3D scene, and the basic monitoring parameters described in S2 are extracted through manual measurement. Next, the distribution of fences, counterweights, structural components, and other debris in the beam surface area is understood within the 3D scene: A schematic diagram of the beam surface area in this embodiment is shown below. Figure 3 As shown, guardrails were installed at the beam ends CD and EF sections on both sides of the beam, and promotional slogans were laid in the diagonally filled area. Based on avoiding interference, suitable installation locations for the BeiDou positioning equipment were selected at points A and B, each 2.5 meters from the beam end, taking into account the structural characteristics. The attitude sensor was installed at point O, located at the center of the rotating beam. The three-dimensional coordinates of points A, B, and O were obtained through measurement, and the BeiDou high-precision positioning equipment was used for layout and installation on site. During the actual rotation process, the real-time attitude monitoring results of the bridge rotation obtained through S4 and S5, along with the monitoring indicators for early warning, were simultaneously displayed on a webpage, showing the current attitude of the rotating beam and real-time early warning information. The visualization interface of the bridge rotation system is shown below. Figure 4 As shown.

[0121] From the formal rotation process of this embodiment, the attitude monitoring results of the Beidou positioning device, the monitoring results of the attitude sensor, and the monitoring results after data fusion are selected over a period of time. Figure 5As shown, the attitude monitoring results of Beidou positioning device A fluctuated significantly during the rotation process. Referring to the original monitoring data, it was determined that the device's signal was interfered with during this period, preventing the acquisition of a fixed solution and thus affecting monitoring accuracy. The fusion results, based on the fusion formula in S412, eliminated the monitoring results from when the device was not fixed, ensuring monitoring accuracy. Therefore, the monitoring and early warning method proposed in this invention can effectively improve the monitoring accuracy and stability during the rotation construction process.

[0122] The measured rotation angle-time curve and the simulated curves at the highest and lowest rotation angular velocities are shown below. Figure 6 As shown: The uniform rotation phase lasted from approximately 15 minutes to 61 minutes, with brief pauses around 23 and 27 minutes. During these adjustments, a low-speed warning was triggered based on the S63 monitoring and early warning formula. During the uniform rotation, the actual rotation curve remained between the low-speed and high-speed curves, without generating any low-speed or high-speed warnings. At approximately 19 minutes into the rotation, with a rotation angle of approximately 9 degrees, the rotating beam entered the railway clearance zone. The deceleration phase lasted from 61 minutes to 68 minutes. At approximately 66 minutes into the rotation, the distance to the target angle was less than 2 degrees, triggering a near-target warning. The fine-tuning phase began at 68 minutes. At approximately 72 minutes into the rotation, the distance to the target angle was less than 0.5 degrees, triggering an over-rotation warning. Finally, through several minor adjustments, the rotation angle reached the target angle. Throughout the entire formal rotation process, all other indicators remained within the normal range, and no tilting or clearance warnings were generated.

[0123] Example 2

[0124] This embodiment applies to a highway swing bridge spanning an existing railway, with a bridge deck structure being a curved beam with constant curvature. The total length of the beam is 120 meters, the curve radius is 2550 meters, and the swing beam rotates to the right from the lower kilometer marker towards the higher kilometer marker, with the rotation center located 15 centimeters to the right of the midpoint kilometer marker. The bridge rotation angle is 64.7 degrees, with a trial rotation of 5 degrees and a final rotation of 59.7 degrees.

[0125] The monitoring process for the rotation construction of curved beams is basically the same as that for straight beams. The main difference is that the schematic diagram of the beam surface area of ​​a curved beam is as follows: Figure 7 As shown, the installation points A and B of the Beidou positioning equipment are both 2.25 meters away from the beam end; the installation point O of the attitude sensor is located at the rotation center. Substituting this installation point information into S412's α... constA The calculation formula yields the corrected constants for the azimuth angles of the rotating beam from Beidou positioning devices A and B when the rotating beam is a curved beam. This method enables precise monitoring of the curved beam's rotational attitude, providing support for real-time visualization and construction control of the rotation.

[0126] This embodiment extracts a segment of the uncorrected azimuth-time curves of the rotating beam from Beidou positioning devices A and B, as well as the corrected and data-fused azimuth-time curves of the rotating beam, as shown below. Figure 8 As shown, the azimuth angle of the rotating beam calculated using devices A and B alone has an error of approximately 0.48 degrees compared to the corrected and merged azimuth angle. If this error is not corrected, it could lead to accidents such as over-rotation of the beam. Therefore, the real-time monitoring and early warning method for railway bridge rotation of this invention can also achieve good monitoring results for curved beam rotation.

[0127] The above description is a detailed description of the preferred embodiments of the present invention. However, the embodiments are not intended to limit the scope of the patent application of the present invention. All equivalent changes or modifications made under the technical spirit of the present invention should fall within the patent scope covered by the present invention.

Claims

1. A method for real-time monitoring and early warning of railway bridge rotation based on multi-source sensing fusion, characterized in that, Includes the following steps: S1 generates a realistic 3D model of the construction site of the railway bridge rotation project; S2, load the real-world 3D model obtained in S1 into the 3D scene, and extract the parameters required for real-time monitoring and early warning; S3, using the real-scene 3D model to analyze the beam surface area of ​​the railway bridge, and combining the structural characteristics of the beam surface, select the installation location for deploying Beidou positioning equipment and attitude sensors; S4 analyzes the monitoring data collected by the Beidou positioning equipment and attitude sensors to obtain key indicator values ​​for the rotation monitoring of railway bridges, including the azimuth angle, tilt angle, instantaneous angular velocity, and average angular velocity of the rotating beam; and performs data fusion on the monitoring results of the Beidou positioning equipment and attitude sensors based on the mean error to obtain high-precision real-time attitude monitoring results of the bridge rotation. S5: Calculate the real-time coordinates of the four corner points of the railway bridge beam based on the real-time attitude monitoring results obtained in S4; monitor whether the outline of the railway bridge intersects with the railway clearance through vector product; if there is an intersection, calculate the clearance from the rotating beam to the top of the contact wire pole; calculate the rotation angle-time simulation curve. S6 constructs a three-dimensional real-world scene of the construction site that integrates BIM and GIS, and drives the dynamic update of the rotating beam's posture by combining real-time monitoring results, thus fully restoring the structural state changes during the rotation process; at the same time, it identifies and warns of abnormal working conditions in real time based on the real-time posture monitoring results and indicators of the bridge rotation in S4 and S5. S4 includes the following sub-steps: S41, Beidou positioning equipment monitoring data processing: The data messages transmitted by the Beidou positioning equipment adopt the standard GPGGA format; the message content includes the equipment number, data measurement time, longitude, latitude, elevation, positioning type, and standard error. After obtaining the data messages in the above format, coordinate transformation and rotation attitude calculation are performed on them. S42, Attitude Sensor Monitoring Data Processing: By parsing the monitoring data messages collected by the attitude sensor, the azimuth angle, tilt angle, instantaneous angular velocity, and average angular velocity of the rotating beam monitored by the attitude sensor are obtained, and effective attitude information parameters are extracted and used as the monitoring results of the attitude sensor; among which: The monitoring data collected by the attitude sensor includes: the rotation angles of the three coordinate axes relative to the initial attitude of the device, the instantaneous rotational angular velocity of each coordinate axis, and the corresponding data acquisition timestamps. This monitoring data is transmitted to a designated server in the form of TCP packets. The server, based on the data packet format and decoding rules, parses and processes the monitoring data frame by frame to extract valid attitude information parameters. As a monitoring result of the attitude sensor; S43, Monitoring Result Fusion: Based on the mean square error of the Beidou positioning equipment and attitude sensor, the monitoring results of the two are weighted and fused to obtain high-precision real-time attitude monitoring results of the bridge rotation, as follows: The fusion process first pairs the monitoring results of BeiDou positioning devices and attitude sensors at the same time point based on the monitoring time, and then performs weighted fusion, as shown in the following formula: In the formula, The weights for the monitoring results from the BeiDou positioning equipment and the attitude sensor are respectively calculated from the mean square error (RMS) of the two types of monitoring results; The final attitude monitoring result is obtained by fusing the monitoring results from the BeiDou positioning equipment and the attitude sensor. The azimuth angle of the rotating beam at the current moment. This represents the change in the beam's tilt angle. It is the instantaneous angular velocity. The average angular velocity; The weights of the two types of monitoring results are calculated based on the mean square error values ​​of the measurement data from the BeiDou positioning equipment and attitude sensor: ， In the formula, These are the weights assigned to the monitoring results from the BeiDou positioning equipment and the attitude sensor, respectively. These are the mean square error values ​​of the measurement data from the Beidou positioning device and the attitude sensor, respectively. S41 includes the following steps: S411, Coordinate Transformation: The geographic coordinate system result of BeiDou positioning is transformed to the construction coordinate system. The coordinate transformation from the geographic coordinate system to the construction coordinate system is calculated using a seven-parameter coordinate transformation model, as shown below: ， In the formula, and These are coordinates in the geographic coordinate system and the construction coordinate system, respectively. These are translation parameters; These are rotation parameters; Scale factor; The translation parameters, rotation parameters, and scale factor in the above seven-parameter coordinate transformation model are calculated using the least squares method based on three or more known control points. S412, Rotation Attitude Calculation: Based on the coordinates of the transformed construction coordinate system obtained in S411, the azimuth angle, change in tilt angle, instantaneous angular velocity, and average angular velocity of the rotating beam are calculated. Data fusion of monitoring results from both large and small mileage terminals is then performed using the positioning type and error of the BeiDou positioning equipment. Details are as follows: The center of rotation of the beam surface extracted from S2 The geographic coordinates of the point are processed using the S411 coordinate transformation model to obtain the construction coordinate system coordinates of the beam surface rotation center. Combined with the initial coordinates of the Beidou positioning devices at both ends of the beam and The initial azimuth angle of the beam was calculated. As shown in the following formula: ， For BeiDou positioning device A, assuming... The construction coordinates of the location of the Beidou positioning equipment are as follows: At this moment, the azimuth angle of the rotating beam And the rotating beam has rotated at an angle The following formulas are used respectively: ， ， In the formula, To correct the azimuth angle corresponding to the shape of the rotating beam to a constant, use the following formula: ， Initial tilt of the bridge , The tilt of time and the change in tilt angle The formula is shown below: ， ， ， Instantaneous angular velocity of rotation With average angular velocity The calculation is shown in the following formula: ， ， Two sets of rotation attitude monitoring results were obtained from two Beidou positioning devices installed at the large and small mileage ends of the beam. These results were then fused using the mean square error. ， In the formula, These represent the attitude monitoring results of data fusion from two BeiDou positioning devices, the attitude monitoring results of BeiDou positioning device A, and the attitude monitoring results of BeiDou positioning device B, respectively. These represent the horizontal mean square errors of Beidou positioning devices A and B, respectively. These represent the positioning types of BeiDou positioning devices A and B, respectively. The definition is as follows: 。 2. The method for real-time monitoring and early warning of railway bridge rotation based on multi-source sensing fusion according to claim 1, characterized in that, S5 includes the following steps: S51, Monitoring of the Rotation Beam Entering the Railway Clearance: Based on the real-time attitude monitoring results, calculate the real-time coordinates of each corner point of the beam, and monitor whether the outline of the rotation beam intersects with the railway clearance through vector product, thereby determining whether the rotation beam has entered the airspace above the railway clearance. S52, Clearance from the top of the contact wire pole to the rotating beam: Based on the fact that the rotating beam has already entered the railway clearance in S51, the half-plane method is used to determine whether the rotating beam passes over the top of the contact wire pole: First, based on the real-time three-dimensional coordinate data of the beam corner points obtained in S51 during the rotation process, and combined with the coordinates of the top of the contact wire pole extracted in S2, the beam is used in a unified coordinate system to determine whether it passes over the top of the contact wire pole using the half-plane method. The half-plane method is used to calculate the vector product of the edge vector and the vector from the contact wire pole to the edge starting point for each edge of the beam. If all the vector products are in the same direction, it means that the beam has passed over the top of the contact wire pole. S53, Calculate the rotation angle-time simulation curve: In order to realize real-time monitoring and early warning of overtime and overspeed during the bridge rotation process, it is necessary to calculate the simulation curve of the beam angle changing with time during the rotation process, compare it with the rotation curve in the actual rotation construction process, and provide data support for the rotation construction control.

3. The method for real-time monitoring and early warning of railway bridge rotation based on multi-source sensing fusion according to claim 2, characterized in that, S51 includes the following steps: S511, based on the initial coordinates of each corner point of the beam extracted in S2. The real-time coordinates are calculated using the following formula: ， In the formula, These are the real-time coordinates of the corner points of the beam. These are the initial coordinates of the corner points of the beam; The coordinates of the rotation center of the rotating beam; This is the difference between the real-time azimuth angle and the initial azimuth angle of the beam. S512, combined with the railway clearance point sequence obtained from S2 The vector product is used to detect whether there is any intersection between each side of the beam and each section of the railway clearance.

4. The method for real-time monitoring and early warning of railway bridge rotation based on multi-source sensing fusion according to claim 2, characterized in that, S52 includes the following steps: S521, When the beam passes over the contact wire pole, the planar coordinates of the top of the contact wire pole are projected onto the centerline of the bridge to obtain the projection point. and corresponding mileage Then the projection point of the beam surface The elevation at point A was obtained by interpolating the beam surface elevations at points A and B using Beidou positioning equipment. The calculation formula is as follows: ， In the formula, These are the mileages at the installation locations of Beidou positioning devices A and B, respectively. The elevation of the beam surface at points A and B of the Beidou positioning equipment; S522, Based on the beam surface projection points obtained in S521 elevation Elevation obtained from S2 Then, combined with the formula for the change of beam thickness with mileage Obtain the clearance from the rotating beam to the top of the contact wire pole. as follows: ， The formula for the change of beam thickness with mileage The sequence of sampling points for mileage and beam surface thickness obtained from S2 at 5-meter intervals. The shape of the beam is obtained by multi-curve fitting using piecewise functions.

5. The method for real-time monitoring and early warning of railway bridge rotation based on multi-source sensing fusion according to claim 2, characterized in that, In S53, the rotation process includes four stages: acceleration, constant speed, deceleration, and fine adjustment, wherein: The acceleration phase begins when the rotation window is opened and ends when the rotation angle reaches 5 degrees. This phase usually requires the safety personnel to go online, the overhead contact line to be powered off, and the rotation equipment to be adjusted. The uniform speed phase begins when the angle of rotation is 5 degrees and ends when the remaining angle is 5 degrees, during which the rotating beam maintains a uniform speed of rotation. The deceleration phase begins when the remaining angle is 5 degrees and ends when the remaining angle is 1 degree, during which the rotation speed gradually decreases until it stops. The fine-tuning phase lasts from when the remaining angle is 1 degree until the rotating beam is fully in place. The rotating beam is positioned by performing multiple inching operations. The rotation angle-time simulation curve is calculated according to the following formula when entering the uniform velocity stage: ， In the formula, The simulated rotation angles during the uniform and deceleration phases are given; t is the rotation time. These represent the times at the start of the uniform velocity phase, the start of the deceleration phase, and the end of the deceleration phase, respectively. The angular velocity during uniform rotation; The angle at which the rotating beam will rotate; This refers to the angular acceleration during the deceleration phase. Based on the above formula, the curves of the rotation angle changing with time under the conditions of rotation construction at the highest and lowest angular velocities are calculated. This allows for early warning of actual rotation curves exceeding these speed ranges. Specifically: The highest rotational angular velocity Calculation based on the cantilever length of the rotating beam: ， In the formula, This is the maximum rotational angular velocity, expressed in rad / min. This is the cantilever length of the rotating beam, which is half the total length of the rotating beam in a symmetrical rotating beam, and is expressed in meters (m). The rotational angular acceleration during the deceleration phase Referring to national and local standards, control within the following; For the minimum rotational angular velocity First, calculate the expected end time of the deceleration phase based on the end time of the rotation window and the reserved time of the fine-tuning phase, and then solve it using the rotation angle-time simulation curve formula.

6. The method for real-time monitoring and early warning of railway bridge rotation based on multi-source sensing fusion according to claim 1, characterized in that, The parameters mentioned in S2 include: beam surface rotation center. geographic coordinates of the point Initial coordinates of the four corner points of the beam Railway line clearance point sequence 3D coordinate sequence of the top of the contact wire pole 5-meter intervals for sampling points of mileage and beam surface thickness , Bridge rotation direction, rotation angle of the rotating beam Beam length Beam width Beam height .

7. The method for real-time monitoring and early warning of railway bridge rotation based on multi-source sensing fusion according to claim 1, characterized in that, In S3, one Beidou positioning device is installed at each of the large and small mileage ends of the beam, and both are installed on fixed supports in an open and unobstructed environment. The installation points of the two Beidou positioning devices are on the centerline of the bridge and are equidistant from the rotation center of the beam surface.

8. The method for real-time monitoring and early warning of railway bridge rotation based on multi-source sensing fusion according to claim 1, characterized in that, In S1, the drone's flight path is parallel to the railway line, and its projection is 25 to 30 meters away from the horizontal distance of the rails. The flight path height is 20 to 30 meters higher than the rotating beam. In S6, model modification includes removing the rotating beam and its attached structures, as well as the surrounding truck crane, from the real-world 3D model.

Images