Novel elevation and level measuring method for laser level tracker

Through the new laser level tracker and deep learning intelligent algorithm, the problem that traditional reservoir dam monitoring methods cannot achieve automation and high precision is solved, and high-precision automatic monitoring of surface displacement of reservoir dams is realized, reducing system complexity and cost .

CN118913206AInactive Publication Date: 2024-11-08BEIJING LAT LASER PRECISION INSTR CO LTD
View PDF 0 Cites 5 Cited by

Patent Information

Application Number
CN202410962996.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-18
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The monitoring methods of traditional reservoir dams cannot achieve automated, all-weather, real-time, unattended high-precision monitoring, especially the horizontal displacement measurement accuracy is difficult to meet the requirements, and the system is complex and costly.

Method used

Using a new laser level tracker combined with deep learning intelligent algorithms, high-precision automated monitoring is achieved through laser ring grid, liquid compensation and ring grid recognition technology, and using GhostNet-YOLOv8-pose improved spot center positioning algorithm to improve monitoring accuracy and efficiency , reduce system complexity and cost.

Benefits of technology

It realizes high-precision automated monitoring of surface displacement of reservoir dams, reduces system complexity and cost, adapts to different environments and scenarios, and meets the needs of automation monitoring.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118913206A_ABST
    Figure CN118913206A_ABST
Patent Text Reader

Abstract

The invention discloses a novel laser level tracker elevation and level measurement method, and in the aspect of elevation measurement, laser level tracker elevation measurement comprises the following steps: step 1, the elevation Ha of a dam level base point A and the elevation Hb of a level base point B are known; the system has the beneficial effects that the monitoring station and the light source station are exquisite in structural design, multiple sensors are integrated in a limited space, and elevation, levelness and temperature parameters can be accurately measured at the same time; the high-precision rotary table is adopted for horizontal displacement monitoring, a satellite positioning system does not need to be adopted, the system complexity is greatly reduced, and the cost is greatly reduced; an intelligent image processing algorithm based on deep learning is adopted as an extraction algorithm of the light spot center position, the system precision is greatly improved, and the system can adapt to different environments and scenes.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of digital automatic monitoring of reservoir dams, and in particular to a novel elevation and level measurement method of a laser level tracker. Background Art

[0002] The elevation values ​​of each monitoring point on the top surface of the main dam, auxiliary dam, spillway and other structural structures of the reservoir dam change with time in the vertical direction relative to the elevation value of the working base point. At the same time, they also have horizontal displacement changes, which require long-term monitoring.

[0003] The traditional observation method is to use precise optical leveling to establish a vertical displacement monitoring and control network through reference points, working base points, and displacement measurement points. This method is scientific and has reliable monitoring accuracy, but it still cannot get rid of manual observation and cannot achieve the requirements of automated, all-weather, real-time, and unattended automatic monitoring. In addition, the horizontal displacement measurement accuracy is usually difficult to meet application requirements and is difficult to transmit. Therefore, horizontal displacement positioning is usually achieved through a satellite positioning system, which makes the system too complex and costly. Summary of the invention

[0004] The purpose of the present invention is to provide a novel laser level tracker elevation and level measurement method in order to solve the above problems.

[0005] The present invention achieves the above-mentioned purpose through the following technical solutions:

[0006] New laser level tracker elevation and horizontal measurement method. In terms of elevation measurement, the laser level tracker elevation measurement includes the following steps:

[0007] Step 1: The elevation of the dam leveling base point A is known to be Ha, and the elevation of the leveling base point B is known to be Hb;

[0008] Step 2: After the cement pier is built, use a high-precision level to measure the vertical displacement monitoring working base points 0# and 4# and monitoring points 1#, 2#, and 3# from the known leveling base points A and B. The elevations of the leveling marks at the bottom of the cement pier are H0 and H4 and H1, H2, and H3, in meters. H1 and H4 are generally unchanged and can be re-measured regularly from the leveling base points A and B.

[0009] Step 3: When installing the equipment for the first time, while completing the elevation measurement in step 2 above, use a high-precision level to measure the heights h0, h1, h2, h3, h4 from the level mark at the bottom of 0#, 1#, 2#, 3#, and 4# piers to the top platform of the piers (in meters).

[0010] Step 4: During the first installation, the center height z0 of each laser level host and the center height b0 of the matching receiving target zero position must be calibrated to the same height on a horizontal platform before leaving the factory, with an error of less than 0.02mm, that is, z0=b0, z1=b1, z2=b2, z3=b3, in mm;

[0011] Step 5: When the laser receiving target collects images and calculates the coordinates, multiple images are required. First, the abnormal image data of the spot is eliminated, and then the Laida criterion is applied to eliminate the gross error data;

[0012] Step 6: After the initial installation, the height measurement system needs to be zero-calibrated, that is, each set of laser level i angle correction;

[0013] Step 7: Calculation of monitoring values ​​during operation. The output values ​​of each point of the monitoring pier during operation are the absolute elevation values ​​of each pier top platform. The calculation method of the absolute elevation value outputted by each point is used and compensated by the ambient temperature.

[0014] In terms of horizontal displacement monitoring, the horizontal displacement measurement of the single-point monitoring position is still measured using the measuring target in the high-precision laser level tracker, and the specific measurement steps are as follows:

[0015] Step 1: Select the measurement target 0 value as x0, and the receiving target 1 actually changes from position 1 to position 2. At this time, the light spot of the measuring light source station 1 changes from x10 to x11, which changes by △x1; that is, the receiving target 1 moves by △d1=△x1. If the light source 1 moves by △x0 (measured by target 0), the measuring target 1 moves by △x0+△x1 relative to target 0.

[0016] Step 2: At this time, the light spot of the measuring light source station 1 changes from x10 to x11, which changes by △x1; that is, the receiving target 1 moves by △d1=△x1. If the light source 1 does not move, the measuring target 1 moves by △x1 relative to the target 0; if the light source 1 moves, the measuring target 1 moves by △x0+△x1 relative to the target 0;

[0017] Step 3: At this time, the light spot of the measuring light source station 2 changes from x20 to x21, which changes by △x2; let the front and rear viewing angle of the light source station 2 be θ2 (approximately equal to 180°); the measurement value of the measuring target 2 is x2. If the light source 2 moves and the measuring target 2 does not move, the measuring target 1 moves △x0+△x1+△x2 relative to the target 0;

[0018] Step 4: Similarly, the information can be transmitted downward without limit, thus realizing the horizontal displacement transmission of all monitoring stations.

[0019] Furthermore, the laser digital level tracker consists of a light source station, a monitoring station and corresponding power supply, communication, control and other components. It adopts three core technologies, namely laser ring grating, liquid compensation, and ring grating identification, to solve the problems of zoom caused by lens refraction in measurement (relative measurement) of traditional vacuum pipeline collimated laser, light spot drift at the laser receiving end, difficulty in ensuring pipeline vacuum pressure, complex system, and high engineering cost, and realizes high-precision automatic monitoring of surface displacement.

[0020] Furthermore, the monitoring station will adopt a laser digital level tracker based on a high-precision turntable. Based on the original surface deformation observation point elevation measurement, the high-precision turntable will be used as a reference to simultaneously realize the horizontal and vertical displacement of the surface deformation observation points to meet the automated monitoring requirements of regulations and specifications.

[0021] Furthermore, in terms of elevation measurement, a laser leveling laser light source and a camera in the monitoring target are used to obtain elevation changes, and the baseline is used as a horizontal reference to transfer the elevation.

[0022] Furthermore, the measurement of horizontal displacement uses a high-precision turntable as a reference, wherein the horizontal displacement measurement of a single-point monitoring position is still measured using the measuring target in a high-precision laser level tracker.

[0023] Furthermore, the image center extraction of the laser digital level tracker is realized through the intelligent algorithm of deep learning. The spot image data required for detection is collected by laser collimation technology, and the spot center is located based on the spot image digital processing technology. An improved spot center positioning algorithm based on YOLOv8n is proposed - GhostNet-YOLOv8-pose. GhostNet-YOLOv8-pose introduces the lightweight module GhostNet to reduce the number of model parameters while ensuring accuracy, thereby improving the speed and efficiency of spot center positioning.

[0024] Furthermore, a digital camera is used in the monitoring station to measure the displacement of the spot image, and the displacement is obtained by extracting the center of the laser spot. Therefore, the extraction of the spot center plays a vital role in the monitoring system. Accurately extracting the spot center can significantly improve the accuracy and efficiency of monitoring. Although the traditional method has achieved certain results in some application scenarios, it has certain limitations in complex environments. The present invention uses a deep learning intelligent algorithm to collect the spot image data required for detection by laser collimation technology, and locates the spot center based on the spot image digital processing technology. A spot center positioning algorithm based on the improved YOLOv8n is proposed—GhostNet-YOLOv8-pose.

[0025] The beneficial effects of the present invention are:

[0026] 1. The monitoring station and light source station in the present invention are exquisitely designed in structure, and multiple sensors are integrated in a limited space, which can accurately measure the elevation, level and temperature parameters at the same time;

[0027] 2. The present invention uses a high-precision turntable to monitor horizontal displacement, without the need for a satellite positioning system, which greatly reduces system complexity and costs;

[0028] 3. The algorithm for extracting the center position of the light spot in the present invention adopts an intelligent image processing algorithm based on deep learning, which greatly improves the system accuracy and can adapt to different environments and scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A schematic diagram of a monitoring system for the novel laser level tracker elevation and level measurement method of the present invention;

[0030] Figure 2 It is a schematic diagram of the horizontal transmission principle scene 1 of the new laser level tracker elevation and level measurement method described in the present invention;

[0031] Figure 3 This is a schematic diagram of the second horizontal transmission principle scenario of the new laser level tracker elevation and horizontal measurement method described in the present invention. DETAILED DESCRIPTION

[0032] A new type of laser level tracker elevation and horizontal measurement method, in which the laser digital level tracker is composed of a light source station, a monitoring station and corresponding power supply, communication, control and other components. It adopts three core technologies, including laser ring grating, liquid compensation, and ring grating recognition, to solve the problems of zoom caused by lens refraction in the measurement (relative measurement) of traditional vacuum pipeline collimated laser, laser receiving end spot drift, difficult to ensure pipeline vacuum pressure, complex system, and high engineering cost, and realizes high-precision automatic monitoring of surface vertical displacement; the communication between the laser digital level tracker and the sub-control center adopts wired communication, and data collection, calculation, analysis and other tasks are all completed on the server computer; the monitoring system will use a laser digital level tracker based on a high-precision turntable, and on the basis of the original surface deformation observation point elevation measurement, use a high-precision turntable as a reference, and realize the horizontal and vertical displacement of the surface deformation observation point at the same time, to meet the requirements of regulations and specifications. Automatic monitoring.

[0033] In terms of elevation measurement, the laser level tracker elevation measurement includes the following steps:

[0034] Step 1: The elevation of the dam leveling base point A is known to be Ha, and the elevation of the leveling base point B is known to be Hb;

[0035] Step 2: After the cement pier is built, use a high-precision level to measure the vertical displacement monitoring working base points 0# and 4# and monitoring points 1#, 2#, and 3# from the known leveling base points A and B. The elevations of the leveling marks at the bottom of the cement pier are H0 and H4 and H1, H2, and H3, in meters. H1 and H4 are generally unchanged and can be re-measured regularly from the leveling base points A and B.

[0036] Step 3: When installing the equipment for the first time, while completing the elevation measurement in step 2 above, use a high-precision level to measure the heights h0, h1, h2, h3, h4 from the level mark at the bottom of 0#, 1#, 2#, 3#, and 4# piers to the top platform of the piers (in meters).

[0037] Step 4: During the first installation, the center height z0 of each laser level host and the center height b0 of the matching receiving target zero position must be calibrated to the same height on a horizontal platform before leaving the factory, with an error of less than 0.02mm, that is, z0=b0, z1=b1, z2=b2, z3=b3, in mm;

[0038] Step 5: When the laser receiving target collects images and calculates the coordinates, multiple images are required. First, the abnormal image data of the spot is eliminated, and then the Laida criterion is applied to eliminate the gross error data;

[0039] Step 6: After the first installation, the height measurement system needs to be zero-calibrated, that is, each set of laser level i angle correction, the correction steps are as follows:

[0040] 1) First calculate the elevation difference Δh between the 1# pier top platform and the 0# pier top platform 10 =(H1+h1)-(H0+h0);

[0041] 2) Then use laser leveling to measure the elevation difference Δy between the 1# pier top platform and the 0# pier top platform 10 =y1-y0, where y1 is the measured value of the receiving target on pier #1, and y0 is the coordinate value of the receiving target on pier #1 when it is calibrated to b0 at the factory;

[0042] 3) Then the i angle correction value of the 0# main engine is i0 = Δh 10 -Δy 10 , if i0 is positive, it means the laser optical axis is above the horizontal plane, otherwise it is negative;

[0043] 4) Use the above method to calibrate i1, i2, and i3 of 1#, 2#, and 3# hosts simultaneously;

[0044] Step 7: Runtime monitoring value calculation

[0045] 1) During operation, the output values ​​of each point of 0#, 1#, 2#, 3#, and 4# piers are the absolute elevation values ​​of each pier top platform. The calculation method of the absolute elevation value output at each point is

[0046] 0# Elevation Z0=H0+h0

[0047] 1# elevation Z1=Z0+Δy 10 +i0

[0048] 2# elevation Z2=Z1+Δy 21 +i1

[0049] 3# elevation Z3=Z2+Δy 32 +i2

[0050] 4# elevation Z4=Z3+Δy 43 +i3

[0051] If H0 and H4 are changed in the later re-measurement, manually input the re-measurement elevation and enter the program parameter setting interface for modification;

[0052] 2) Correction of the impact of ambient temperature changes

[0053] According to the actual measurement of the laser optical axis with the ambient temperature change, the 0#, 1#, 2#, 3# laser level correction coefficients k0, k1, k2, k3 and the 1#, 2#, 3#, 4# pier elevation corrections m1, m2, m3, m4 are calculated. The temperature coefficient k is automatically calculated once a month based on the original measurement value to ensure the accuracy of the coefficient.

[0054] The elevation of each point is calculated as follows:

[0055] 0# Elevation Z0=H0+h0

[0056] 1# elevation Z1=Z0+Δy 10 +i0+m1

[0057] 2# elevation Z2=Z1+Δy 21 +i1+m2

[0058] 3# elevation Z3=Z2+Δy 32 +i2+m3

[0059] 4# elevation Z4=Z3+Δy 43 +i3+m4

[0060] 3) Calculation of height difference closure error

[0061] The measured elevation difference between pier 4# and pier 0# is:

[0062] ΔH 40 =Z4-Z0=Δy 43 +i3+m4+Δy 32 +i2+m3+Δy 21 +i1+m2+Δy 10 +i0+m1

[0063] The known elevation difference between pier 4# and pier 0# is:

[0064] ΔH 40 ′=(H4+h4)-(H0+h0)

[0065] Height difference closure error f h =ΔH 40 -ΔH 40 ′, unit: m;

[0066] 4) Closure error verification and allocation

[0067] There are four measuring stations in this project, so n = 4, the distance between each measuring station is 50m, and the empirical value δ = ±0.0007m

[0068] Closure error allowable error

[0069] If f h ≤f h '

[0070] Then the height difference closure correction number for each measuring station is

[0071] After the closure error correction, the absolute elevation values ​​of each point are calculated as follows:

[0072] 0# Elevation Z0=H0+h0

[0073] 1# elevation Z1=Z0+Δy 10 +i0+m1+v

[0074] 2# elevation Z2=Z1+Δy 21 +i1+m2+v

[0075] 3# elevation Z3=Z2+Δy 32 +i2+m3+v

[0076] 4# elevation Z4=Z3+Δy 43 +i3+m4+v;

[0077] In terms of horizontal displacement monitoring, the horizontal displacement measurement of the single-point monitoring position is still measured using the measuring target in the high-precision laser level tracker, and the specific measurement steps are as follows:

[0078] Step 1: In the figure, the value of the measuring target 0 is x0, and the receiving target 1 actually changes from position 1 to position 2. At this time, the light spot of the measuring light source station 1 changes from x10 to x11, which changes by △x1; that is, the receiving target 1 moves by △d1=△x1. If the light source 1 moves by △x0 (measured by target 0), the measuring target 1 moves by △x0+△x1 relative to target 0.

[0079] Step 2: At this time, the light spot of the measuring light source station 1 changes from x10 to x11, which changes by △x1; that is, the receiving target 1 moves by △d1=△x1. If the light source 1 does not move, the measuring target 1 moves by △x1 relative to the target 0; if the light source 1 moves, the measuring target 1 moves by △x0+△x1 relative to the target 0;

[0080] Step 3: At this time, the light spot of the measuring light source station 2 changes from x20 to x21, which changes by △x2; let the front and rear viewing angle of the light source station 2 be θ2 (approximately equal to 180°); the measurement value of the measuring target 2 is x2. If the light source 2 moves and the measuring target 2 does not move, the measuring target 1 moves △x0+△x1+△x2 relative to the target 0;

[0081] Step 4: Similarly, the information can be transmitted downward without limit, thus realizing the horizontal displacement transmission of all monitoring stations.

[0082] In addition, further explanation is needed, such as Figure 3 As shown in the figure, when the front and rear viewing angles of the light source station are not 180 degrees, the algorithm is still valid. It only needs to calibrate the specific angle difference between the front and rear viewing angles in advance and bring it into the solution.

[0083] A digital camera is used in the monitoring system to measure the displacement of the spot image, and the displacement is obtained by extracting the center of the laser spot. Therefore, the extraction of the spot center plays a vital role in the monitoring system. Accurately extracting the spot center can significantly improve the accuracy and efficiency of monitoring. Although the traditional method has achieved certain results in some application scenarios, it has certain limitations in complex environments. The present invention uses a deep learning intelligent algorithm to collect the spot image data required for detection by laser collimation technology, and locates the spot center based on the spot image digital processing technology.

[0084] An improved light spot center positioning algorithm based on YOLOv8n, GhostNet-YOLOv8-pose, is proposed. GhostNet-YOLOv8-pose introduces the lightweight module GhostNet to reduce the number of model parameters while ensuring accuracy, thereby improving the speed and efficiency of light spot center positioning. In the light spot image data set, compared with the YOLOv8n-pose model, the GhostNet-YOLOv8-pose model ensures higher accuracy and stability in various complex scenarios, meeting the monitoring system's needs for light spot center positioning detection.

[0085] To realize intelligent core extraction of deep learning, it is necessary to establish a data set and extract accurate key frames. The present invention proposes a key frame extraction method based on sampling. Key frame extraction based on sampling is a simple and direct method. It sets an interval value manually and then selects a frame as a key frame at this fixed interval. The advantage of this method lies in its simplicity and robustness, because it is not affected by environmental conditions and the speed of scene changes. However, this method also has obvious disadvantages. First, the number of key frames extracted is directly affected by the set interval value. If the interval value is too small, it may lead to the extraction of too many invalid frames. Second, the key frames extracted by this method may lack representativeness, because most of the extracted frames may be repeated images and information, and the most critical image information may be missed.

[0086] Data augmentation is a common method to improve model performance, and it is also applicable to target detection tasks. Data augmentation obtains new training data by performing a series of transformations on the original image, thereby effectively increasing the amount of data and expanding the training data set. Data augmentation can generate different image variants, thereby enriching the training data set, reducing data bias, and alleviating the overfitting problem. In addition, data augmentation can also improve the robustness of the model, enabling it to better handle unknown new images, thereby improving the performance of the model in practical applications.

[0087] In terms of data enhancement, Compose and OneOf are two very useful code calling methods. Compose is a container that combines multiple data enhancement operations into a single operation. This container executes all the transformations contained in it in sequence. This means that a series of enhancement operations can be connected in series to process the image continuously. OneOf provides a randomness that randomly selects one from a given list of transformations to execute. This makes the data enhancement process more unpredictable, helping the model learn more changes and thus improve its generalization ability.

[0088] The parameters and resource consumption of the training model increase, and the detection performance is getting better and better, but it is often accompanied by a large amount of computing and memory requirements, which limits their application on resource-constrained devices. Therefore, the present invention customizes and lightweights the YOLOv8 model, and GhostNetV2 is an enhanced version of GhostNet. It uses deregulated fully connected (DFC) attention to enhance the output characteristics of the Ghost module, thereby capturing the long-term dependencies between pixels in different spaces and further improving the expressiveness of the lightweight model. DFC attention is based on the original feature Z∈R H×W×CAs input, features are aggregated in the horizontal and vertical directions respectively, the calculation of the fully connected layer is reduced and long-range dependencies are inserted. Overall, the GhostNetV2 framework adopts decoupled fully connected attention, which improves the expressiveness of the model while reducing the number of model parameters, achieving a better trade-off between accuracy and speed. The lightweight CPU convolutional neural network PP-LCNet uses the MKLDNN acceleration module, that is, the multi-core learning deep neural network, which is a technology for accelerating deep learning models for multi-core CPUs. The design goal of PP-LCNet is to significantly improve the performance and accuracy of lightweight models on multiple tasks while keeping the latency almost unchanged. With these improvements, the accuracy of PPLCNet can greatly exceed that of previous network structures at the same classification inference time.

[0089] The model is evaluated using a series of indicators, including meanAverage Precision (mAP), the total number of parameters required by the model (Params), the number of frames processed per second (FPS) by the model, and the number of floating-point operations (GigaFloating-point Operations GFLOPs) performed by the model during reasoning. The following are the specific calculation formulas for each indicator:

[0090]

[0091]

[0092] The annotation file output by the selected GhostNet-YOLOv8-pose improved model is in YOLO format. The center of the light spot located by it is the normalized coordinate value. Therefore, we need to convert the normalized y coordinate and x coordinate into pixel coordinates first. The formula is:

[0093] y pixel =y normalized ×image height

[0094] x pixel =x normalized ×image widht

[0095] Because the data enhancement methods used did not change the spot position, the center position of the spot of each group of images detected by the GhostNet-YOLOv8-pose improved model essentially did not change. Therefore, in order to more intuitively display the center position predicted by the model in this experiment, the average x and y coordinate values ​​​​obtained from the 10 images are regarded as the real spot center. The formula is:

[0096]

[0097] Where N is the number of detections.

[0098] The formula for calculating the standard deviation of the change in the center position of the light spot during multiple detections is:

[0099]

[0100] in, and is the average of the x- and y-coordinates from multiple detections.

[0101] Since the positioning accuracy of the light spot is usually expressed as the root mean square deviation (RMSD) of the x-coordinate and y-coordinate of the light spot, the pixel-level positioning accuracy of the detected light spot is calculated in this experiment using the formula:

[0102]

[0103] Among them, σ x and σ y is the standard deviation of the x- and y-coordinates after multiple detections.

[0104] Finally, the positioning accuracy of pixel units is converted into millimeter units according to the actual physical size of the image taken by the monitoring target camera. The formula is:

[0105] Precision mm =Precision pixel ×Pixel mm

[0106] The combination of these two methods can create very complex and diverse data enhancement pipelines, helping machine learning models better learn and adapt to new situations, thereby improving their performance in real-world applications.

[0107] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A new laser level tracker elevation and level measurement method, characterized by: In terms of elevation measurement, the laser level tracker elevation measurement includes the following steps: Step 1: The elevation of the dam leveling base point A is known to be Ha, and the elevation of the leveling base point B is known to be Hb; Step 2: After the cement pier is built, use a high-precision level to measure the vertical displacement monitoring working base points 0# and 4# and monitoring points 1#, 2#, and 3# from the known leveling base points A and B. The elevations of the leveling marks at the bottom of the cement pier are H0 and H4 and H1, H2, and H3, in meters. H1 and H4 are generally unchanged and can be re-measured regularly from the leveling base points A and B. Step 3: When installing the equipment for the first time, while completing the elevation measurement in step 2 above, use a high-precision level to measure the heights h0, h1, h2, h3, h4 from the level mark at the bottom of 0#, 1#, 2#, 3#, and 4# piers to the top platform of the piers (in meters). Step 4: During the first installation, the center height z0 of each laser level host and the center height b0 of the matching receiving target zero position must be calibrated to the same height on a horizontal platform before leaving the factory, with an error of less than 0.02mm, that is, z0=b0, z1=b1, z2=b2, z3=b3, in mm; Step 5: When the laser receiving target collects images and calculates the coordinates, multiple images are required. First, the abnormal image data of the spot is eliminated, and then the Laida criterion is applied to eliminate the gross error data; Step 6: After the initial installation, the height measurement system needs to be zero-calibrated, that is, each set of laser level i angle correction; Step 7: Calculation of monitoring values ​​during operation. The output values ​​of each point of the monitoring pier during operation are the absolute elevation values ​​of each pier top platform. The calculation method of the absolute elevation value outputted by each point is used and compensated by the ambient temperature. In terms of horizontal displacement monitoring, the horizontal displacement measurement of the single-point monitoring position is still measured using the measuring target in the high-precision laser level tracker, and the specific measurement steps are as follows: Step 1: Select the measurement target 0 value as x0, and the receiving target 1 actually changes from position 1 to position 2. At this time, the light spot of the measuring light source station 1 changes from x10 to x11, which changes by △x1; that is, the receiving target 1 moves by △d1=△x1. If the light source 1 moves by △x0 (measured by target 0), the measuring target 1 moves by △x0+△x1 relative to target 0. Step 2: At this time, the light spot of the measuring light source station 1 changes from x10 to x11, which changes by △x1; that is, the receiving target 1 moves by △d1=△x1. If the light source 1 does not move, the measuring target 1 moves by △x1 relative to the target 0; if the light source 1 moves, the measuring target 1 moves by △x0+△x1 relative to the target 0; Step 3: At this time, the light spot of the measuring light source station 2 changes from x20 to x21, which changes by △x2; let the front and rear viewing angle of the light source station 2 be θ2 (approximately equal to 180°); the measurement value of the measuring target 2 is x2. If the light source 2 moves and the measuring target 2 does not move, the measuring target 1 moves △x0+△x1+△x2 relative to the target 0; Step 4: Similarly, the information can be transmitted downward without limit, thus realizing the horizontal displacement transmission of all monitoring stations.

2. The novel laser level tracker elevation and level measurement method according to claim 1 is characterized in that: The laser digital level tracker consists of a light source station, a monitoring station and corresponding power supply, communication, control and other components. It adopts three core technologies, namely laser ring grating, liquid compensation and ring grating identification, to solve the problems of zoom caused by lens refraction in the measurement (relative measurement) of traditional vacuum pipeline collimated laser, light spot drift at the laser receiving end, difficulty in ensuring pipeline vacuum pressure, complex system and high engineering cost, and realizes high-precision automatic monitoring of surface displacement.

3. The novel laser level tracker elevation and level measurement method according to claim 2 is characterized in that: The monitoring station will adopt a laser digital level tracker based on a high-precision turntable. Based on the original surface deformation observation point elevation measurement, the high-precision turntable will be used as a reference to simultaneously realize the horizontal and vertical displacement of the surface deformation observation points to meet the automated monitoring requirements of regulations and specifications.

4. The novel laser level tracker elevation and level measurement method according to claim 1 is characterized in that: In terms of elevation measurement, a laser level laser light source and a camera in the monitoring target are used to obtain the elevation change, and the baseline is used as the horizontal reference to transfer the elevation.

5. The novel laser level tracker elevation and level measurement method according to claim 1 is characterized in that: The measurement of horizontal displacement uses a high-precision turntable as a reference, and the horizontal displacement measurement of a single-point monitoring position still uses the measuring target in a high-precision laser level tracker for measurement.

6. The novel laser level tracker elevation and level measurement method according to claim 1 is characterized in that: The image center extraction of the laser digital level tracker is realized through the intelligent algorithm of deep learning. The spot image data required for detection is collected by laser collimation technology, and the spot center is located based on the spot image digital processing technology. An improved spot center positioning algorithm based on YOLOv8n is proposed, namely GhostNet-YOLOv8-pose. GhostNet-YOLOv8-pose introduces the lightweight module GhostNet to reduce the number of model parameters while ensuring accuracy, thereby improving the speed and efficiency of spot center positioning.

Citation Information

Cited By

  • Settlement monitoring method and system based on bidirectional one-dimensional displacement sensing

    CN121140726A

  • Accurate displacement measuring method of laser level measuring instrument based on data analysis

    CN121297681A

  • Precise displacement measurement method of laser level meter based on data analysis

    CN121297681B

  • Leveling instrument i angle self-adaptive calibration method and system

    CN121577067A

  • A method and system for self-adapting calibration of a level i angle

    CN121577067B